DNA - FAQs

​

Before you start

This page answers common questions about Soundverse DNA. DNA creation has two current experiences:

1. Full Production DNA
2. Voice DNA

Use the detailed DNA Help Center when you need a step-by-step walkthrough. Only upload material that you own or are authorized to use. For Voice DNA, you must also be the speaker or have the speaker’s consent.

​

1. What is DNA?

DNA is a Soundverse creator profile built from audio. It gives Soundverse material to process into available DNA models and track references. Those models and references can then appear in the DNA generation experience. DNA is not a promise that every upload produces the same number or type of results. The results depend on the supplied audio and the processing outcome.

​

2. What is Full Production DNA?

Full Production DNA is the catalog-based DNA flow. Use it for songs, instrumentals, or acapellas that represent a creator’s sound. It builds and updates the creator’s primary DNA profile. Processing can produce music-style models and track references from that catalog. Use this flow when the goal is to represent a broader music catalog.

​

3. What is Voice DNA?

Voice DNA is the voice-focused training flow. It uses clean solo-vocal recordings to train a voice model. It is separate from the Full Production upload and processing flow. In the current primary workflow, the trained voice model is attached to the creator’s primary DNA profile. Voice DNA does not replace the music catalog already associated with Full Production DNA.

​

4. What is the difference between Full Production DNA and Voice DNA?

Full Production DNA learns from music catalog material. Voice DNA learns from recordings of one speaker or singer. Full Production DNA has a sensitivity control for catalog processing. Voice DNA has an epoch-based quality control for voice training. Choose the flow that matches the material you are uploading.

​

5. What is a primary DNA profile?

The primary DNA profile is the current Full Production home for a creator. New catalog uploads and later processing runs are added to that profile. Existing models and tracks stay in place while new material is added. The primary profile is intended to be the continuing home for the creator’s DNA work.

​

6. How many primary Full Production DNA profiles can I create?

The current Full Production flow maintains one primary DNA profile per owner. It does not create a second primary Full Production DNA profile for the same owner. Instead, add more authorized audio to the existing primary profile. Then run processing again when the new upload is ready to process. Older legacy DNA entries can still appear for some accounts. They do not change the one-primary behavior of the current creation flow.

​

7. Can I upload more music after my first Full Production DNA run?

Yes. The current primary flow is designed for later uploads and processing runs. New models and track results are added to the existing primary DNA profile. Your earlier models and tracks are intended to stay in place. This is different from creating a separate primary DNA.

​

8. Who should use Full Production DNA?

Use Full Production DNA when you want to represent a creator’s musical catalog. Good examples include songs, instrumentals, and acapellas that belong to the same creator or sound direction. The upload collection should reflect what you want the DNA to learn. An organized catalog is easier to reason about than unrelated files gathered only to increase the count.

​

9. Who should use Voice DNA?

Use Voice DNA when the goal is to train a voice-focused model. The source material should be recordings of one speaker or singer. Do not mix several people into the same voice-training dataset. Voice DNA is not the right flow for a mixed song catalog with multiple vocalists and backing music.

​

10. Do I need permission to upload audio?

Yes. For Full Production DNA, confirm that you have ownership and authority to upload the material. For Voice DNA, confirm that the voice is yours or that you have the singer’s consent to clone it. Do not use audio merely because it is publicly available online. The upload confirmation is not a substitute for obtaining the rights you need.

​

11. Can I upload someone else’s voice?

Only if you have that speaker’s or singer’s consent. The Voice DNA flow requires a consent confirmation before training can start. Do not treat the presence of a voice recording as evidence of consent. If consent is unclear, do not train the voice.

​

12. Where do I start a DNA?

Open the DNA area in Soundverse. Choose the training option that matches your goal. Use Full Production DNA for a music catalog. Use Voice DNA for solo-vocal voice training. The current Full Production flow opens the primary DNA profile rather than a blank second primary profile.

​

13. What are the Full Production DNA steps?

The current Full Production DNA flow has five steps:

1. Profile Settings
2. Upload Section
3. Process DNA
4. Advanced
5. Finalize

Processing begins from the Process DNA step. The later steps let you review and configure the primary DNA while the catalog work is underway.

​

14. What audio formats can I use for Full Production DNA?

The current Full Production upload control accepts:

• .mp3
• .wav
• .aac
• .ogg
• .flac
• .m4a

Use an accepted audio file instead of an image, archive, document, or video file.

​

15. How many files can I upload for Full Production DNA?

The current Full Production flow accepts at least one audio file. There is no minimum duration requirement in the current upload text. The current maximum is 1,000 audio files. More files are not automatically better. Choose files that represent the sound you want the primary DNA to learn.

​

16. What kind of Full Production files should I choose?

The interface supports songs, instrumentals, and acapellas. Choose material that is relevant to the intended DNA sound. For example, a catalog can contain works that share a creator, style direction, or production identity. Avoid treating the upload as a general file dump. The processing output can vary when the source collection has very different material.

​

17. Is an acapella a Voice DNA recording?

Not automatically. An acapella can be uploaded to the Full Production DNA catalog flow. Voice DNA is specifically for clean solo-vocal recordings used to train a voice-focused model. Use the Voice DNA requirements when the goal is voice training. Use the Full Production rules when the goal is catalog-style processing.

​

18. What does the Full Production ownership confirmation mean?

It is the in-product confirmation that you have the authority to upload the selected music files. Read it carefully before continuing. It applies to each file you include in that Full Production upload. Do not check it for material that you are not authorized to use.

​

19. What happens while a Full Production upload is running?

The Full Production uploader transfers each selected audio file through an upload session. The interface reports upload progress. Keep the file names and the selected files available until the upload has finished. Do not assume that a file is usable for processing until the upload completes successfully.

​

20. Can a Full Production upload resume after a temporary interruption?

The current uploader keeps local checkpoints for a valid upload session. When possible, it can re-read the session and avoid re-sending chunks that were already confirmed. If a temporary error occurs, retry with the same file selection. An expired or invalid upload session may need a new server-issued session. The uploader surfaces non-retryable failures instead of silently pretending that the file completed.

​

21. What should I do if a Full Production upload fails?

First, keep the source files available. Check that each file is an accepted audio format. Retry the upload with the same file selection. If it fails again, note the error text, approximate time, and the file count. Do not start processing until the upload area shows that the files are ready.

​

22. Can I cancel a Full Production upload?

The current upload code supports an explicit cancellation path. Do not rely on closing the browser tab as a documented cancellation workflow. Use the control that the product exposes when a cancel option is available. If you are unsure whether a transfer completed, return to the DNA flow and check the visible upload state before processing.

​

23. What can I edit in Full Production Profile Settings?

The primary profile flow includes fields for the creator name. It also includes a description, tags, and cover art. The current description field is limited to 300 characters. Use a clear name and description so you can recognize the profile later. Tags are useful for organizing the profile in the product.

​

24. Do I need a cover image?

The creation flow provides a cover-art area. You can use the available cover controls when you want the primary DNA profile to have custom artwork. Cover art does not replace the uploaded audio. The audio catalog is still what the Full Production processing flow uses.

​

25. What do Full Production tags do?

Tags are profile metadata shown in the creation flow. Use concise words that help describe or organize the DNA. The app provides suggested tags as well as an input for your own tags. Tags do not guarantee that a particular model, track, or output style will be produced.

​

26. What does the Full Production sensitivity control do?

Full Production DNA uses a sensitivity slider from 0 to 10. The default value is 5. Lower sensitivity is described as producing broader genre DNA results. Higher sensitivity is described as producing more niche style DNA results. Sensitivity changes how the uploaded catalog is processed. It does not create additional primary DNA profiles.

​

27. What is a good Full Production sensitivity to start with?

The current default is 5. The interface labels that middle area as balanced. Use a lower value when you want broader genre-style grouping. Use a higher value when you want more niche style grouping. There is no source-backed rule that one sensitivity value is best for every catalog.

​

28. When does Full Production processing actually start?

Processing starts from the Process DNA step. The app requires uploaded audio that is ready to process. It sends the pending audio identifiers and the selected sensitivity to the processing flow. The Finalize step does not replace the Process DNA start action.

​

29. Why can I not start Full Production processing twice?

The current flow prevents another processing run while DNA processing is already running for that primary DNA. This protects the current model run from being started again immediately. Wait for the active run to finish before starting another run. Return to My DNA to check whether new models or tracks have appeared.

​

30. What happens when I process new Full Production uploads?

The primary flow uses an additive mode for later uploads. It keeps the existing core playlists rather than clearing them before re-clustering. New processing can add results to the primary DNA. It is not a promise that every run creates a new visible style model. Results depend on the processed catalog.

​

31. What happens after Full Production processing completes?

Soundverse can add model and track results to the primary DNA profile. The processing pipeline can classify available audio with song, vocal, or instrumental reference information and style-related tags. Not every catalog produces every possible reference type. Do not expect a fixed number of models, tracks, or tags from every run.

​

32. Will I receive an email when my Full Production model is ready?

Soundverse attempts to send a model-ready email after a completed Full Production model run. That email is best effort. The persisted model result is available to the product before the email attempt can affect it. An email delivery problem should not make a completed model disappear. Check My DNA for the authoritative in-product result.

​

33. Do I need to keep the page open while Full Production DNA processes?

No. Once processing has started, you can leave the page and return later. Use My DNA to review the result. Do not rely only on an email because delivery is best effort.

​

34. What are the Advanced settings in Full Production DNA?

The current Advanced area includes:

• Advanced Options
• Tagging & Categorization
• Review

The exact fields displayed are part of the current creation UI. Review the values carefully before finishing the flow.

​

35. What can I configure under Full Production Advanced Options?

The current UI includes controls for:

• Visibility
• Price
• License
• Track visibility
• Become Partner

These are configuration controls, not a standalone legal or payout guarantee. Read the current in-product labels and applicable terms before using a public or commercial setting.

​

36. What visibility choices does Full Production DNA have?

The current Full Production creation UI offers:

• Public
• Private
• Personal

Use Private or Personal while you are training and testing a DNA. Use Public only when you are ready for the DNA to be considered for public availability.

​

37. Does Public visibility immediately put my DNA in Discover?

No. A DNA must be both Public and Approved before it is eligible for public discovery. Public visibility is necessary but not sufficient. Approval is a separate status requirement.

​

38. What moderation statuses can a DNA have?

The DNA contract includes statuses such as:

• Pending
• Approved
• Rejected
• On hold

These statuses are distinct from visibility. For example, a Public DNA that is still Pending is not eligible for public discovery.

​

39. What does Pending mean for public discovery?

Pending means the DNA has not reached the Approved state required for public discovery. Keep the DNA’s actual status in view when deciding whether it should be public. Do not promise a public link to other people until the product shows that the DNA is eligible. This documentation does not promise a review time or outcome.

​

40. What does Private or Personal mean for Full Production DNA?

Private and Personal are not public discovery states. They are useful while you are setting up, testing, or keeping a DNA outside public discovery. Public discovery still requires Public visibility and Approved status. Use the actual product controls and statuses shown for your DNA.

​

41. Can I set a Voice DNA to Personal?

The current Voice DNA settings step shows private and public visibility options. Do not assume that Voice DNA exposes the same three visibility choices as Full Production DNA. Use the visibility value available in the Voice DNA form.

​

42. What happens in the Full Production review area?

The current Review area has tabs for:

• Models
• Tracks
• Voice Models

It also includes an optional Review Notes field. The current Review Notes input is limited to 500 characters. Review fields and available results are not a guarantee of public approval.

​

43. What are model references?

Model references are available DNA models that can be selected in the generation experience. They represent a broader DNA model starting point when the interface makes them available. Availability depends on the selected DNA and the processing result. Do not assume that every DNA has the same model list.

​

44. What are track references?

Track references are available processed audio references associated with a DNA. The interface can categorize them as song, vocal, or instrumental references. Choosing a track reference narrows the musical starting point for the available generation flow. It does not guarantee that every generation control will be enabled.

​

45. What generation styles does the current DNA contract support?

The current DNA generation contract accepts:

• Song
• Instrumental

The backend does not accept a standalone vocal or acapella generation style. The UI can show vocal references, but that does not create a universal third output style.

​

46. Why is a generation option missing or disabled?

Generation controls are context-aware. They depend on the selected DNA, available model, and selected track reference. Use the controls that are enabled for the current selection. If an option is absent, that selection does not currently support it. Changing to another available model or reference can change the controls shown.

​

47. Can a vocal reference generate a standalone vocal output?

Do not assume that it can. A vocal reference can appear in the interface as a type of track reference. The current generation request contract still maps available generation to Song or Instrumental. Use the enabled product controls rather than expecting a separate universal vocal output.

​

48. What are the Voice DNA steps?

The current Voice DNA flow has five steps:

1. Voice Profile
2. Upload Recordings
3. Settings
4. DNA Quality
5. Finalize

The dedicated Start Training action appears on the DNA Quality step. The Finalize step then shows the training status.

​

49. What should I upload for Voice DNA?

Upload clean solo-vocal recordings. Aim for at least 10 minutes of usable voice audio. Use one speaker or singer only. Avoid background music and heavy effects. The trainer can re-slice the recording, so you do not need to pre-cut every clip.

​

50. What audio formats can I use for Voice DNA?

The current Voice DNA upload control accepts:

• .mp3
• .wav
• .aac
• .ogg
• .flac
• .m4a

It also accepts audio files through the browser audio input.

​

51. Is one Voice DNA recording enough to continue?

The current Voice DNA form requires at least one recording. The product guidance still says to aim for at least 10 minutes of clean solo-vocal audio. One short file may satisfy the minimum form check but may not represent a useful training dataset. Use the guidance, not only the minimum validation, when preparing voice material.

​

52. Can I record audio directly in the Voice DNA flow?

The current Voice DNA upload area includes a Record control. It also lets you select files or drag audio files into the upload area. Browser microphone permissions and recording availability depend on the device and browser. Use a clean recording before beginning training.

​

53. What information can I enter for a Voice DNA profile?

The Voice DNA profile area includes:

• Creator Name
• Description
• Tags
• Cover image or avatar selection

The current description field is limited to 300 characters. Use a name that helps you identify the voice model later.

​

54. Can Voice DNA be attached to an existing profile?

The Voice DNA interface contains a profile step. The current primary mode attaches the voice model to the canonical primary DNA and leaves the existing DNA content unchanged. Use the profile context shown in the current product flow. Do not expect Voice DNA training to replace your catalog models or tracks.

​

55. What does Voice DNA consent require?

Before moving forward from the upload step, the current form requires:

• At least one recording
• The consent confirmation

The confirmation says that the voice is yours or that you have the singer’s consent to clone it. If you cannot truthfully make that confirmation, do not train the voice.

​

56. What is Voice DNA quality?

Voice DNA uses an epoch-based quality control. The range is 10 to 350 epochs. The default is 180 epochs. The current UI describes 40 to 200 epochs as a typical range. Higher values are described as taking longer while aiming for a higher-fidelity clone.

​

57. What should I choose for Voice DNA quality?

Start with the balanced default unless you have a clear reason to change it. The current labels are:

• Fast: 10 epochs
• Balanced: about 180 epochs
• Studio: 350 epochs

The interface also shows an estimated duration and token cost for the selected epoch value. Those estimates are UI estimates, not a completion-time promise.

​

58. When does Voice DNA training start?

Voice training starts when you select Start Training on the DNA Quality step. The app validates the profile and upload requirements before submission. It then moves to Finalize and begins showing the current training state.

​

59. What does the Voice DNA Finalize step show?

Finalize shows a summary of the Voice DNA. It shows the progress bar and status while the model is being prepared or trained. When the model is ready, it provides a way to view it in the profile. If training fails, the UI reports the failure and returns you to the quality step so you can retry.

​

60. What Voice DNA training states can I see?

The current training view can show:

• Queued
• Preparing or processing
• Training progress by epoch
• Completed
• Failed

Queued means the request is waiting for available training capacity. Preparing means the dataset is being processed before epoch updates are available.

​

61. Can I close the Voice DNA window while it trains?

Yes. The current UI states that the training keeps running when you close the window. The trained result appears under the profile’s Voice DNAs when it is ready. Return to the profile to inspect the finished model.

​

62. What should I do if Voice DNA upload fails?

Check that the recordings are accepted audio files. Confirm that at least one recording is selected. Confirm the consent checkbox. Retry the upload after reading the visible error message. If it continues to fail, keep the error text and the approximate time for support.

​

63. What should I do if Voice DNA training fails?

Read the status shown in Finalize. The current UI returns to the DNA Quality step after a start failure so you can retry. Check the voice material, consent confirmation, selected quality, and current product availability. Do not create a new training request repeatedly while a previous request is still showing active progress.

​

64. Where can I manage my DNA after training?

Use My DNA to review your primary DNA, available models, tracks, and voice models. It is the right place to check the in-product result after Full Production processing or Voice DNA training. Use the status shown in the product rather than relying only on a notification.

​

65. What is Discover?

Discover is the public DNA browsing area. Only DNA content that satisfies the product’s public eligibility rules can appear there. For the current Full Production path, that requires Public visibility and Approved status.

​

66. What is Earn?

The Earn area contains the current Stripe Connect flow for creator earnings where it is available. When no Stripe account is connected, it presents a connection action. When an account is connected, it can show the current DNA dashboard summary and a way to open Stripe. Use the in-product dashboard for the information currently available to your account.

​

67. Does connecting Stripe guarantee a payout?

No. Stripe connection is not itself a guarantee of payout eligibility, royalty amount, payout timing, or commercial rights. The current terms, pricing, licenses, eligibility rules, account status, and Stripe requirements all matter. Read the applicable in-product information before making financial or release decisions.

​

68. What do the Full Production price and license controls guarantee?

They configure values presented in the product. They do not independently explain every legal, commercial, or payout effect. Read the current license label and Soundverse terms before releasing, distributing, or commercially using generated work. This FAQ is not legal advice.

​

69. How do I prepare a useful support report for a DNA issue?

Include:

• Whether the issue is Full Production DNA or Voice DNA
• The step where it happened
• The visible error message
• The approximate time and timezone
• The number and formats of the selected files
• Whether the issue is upload, processing, visibility, generation, or earnings

Do not send passwords, API keys, payment details, or private audio unless support specifically provides a secure way to do so.

​

70. What should I check before reporting a missing DNA result?

Check the primary DNA in My DNA. Check whether the upload actually completed. Check whether processing is still active. Check whether you are looking at the intended model or track reference. For public discovery, check both visibility and approval status. For Voice DNA, check whether training reports Queued, Preparing, Training, Completed, or Failed.

​

71. What should I avoid promising to collaborators or listeners?

Do not promise an exact processing duration. Do not promise a specific number of models or tracks. Do not promise public discovery before the DNA is both Public and Approved. Do not promise a payout merely because Stripe is connected. Do not promise a standalone vocal output from every vocal reference.

​

72. Where can I find the full walkthrough?

Use the DNA Help Center for the full end-to-end guide. It covers preparation, each creation step, processing, visibility, generation, Voice DNA, earnings boundaries, and troubleshooting. Return to this FAQ when you need a quick answer to one specific question.

​

More about DNA types and generation

​

73. Is Full Production DNA a voice clone?

No. Full Production DNA is the catalog-based flow for songs, instrumentals, and acapellas that represent a broader musical identity. It can include vocal material as part of a catalog, but it is not the dedicated flow for cloning one particular person’s voice. Use Voice DNA when the goal is to train a model from clean recordings of one speaker or singer.

​

74. Does Voice DNA learn the production style of my songs?

No. Voice DNA is for the voice itself: one speaker or singer’s vocal character from clean solo-vocal recordings. It does not replace the catalog-based Full Production DNA flow for production style, songwriting context, instrumentals, arrangements, or a broader music catalog. Use both flows only when you have the required rights and each source set matches the intended training purpose.

​

75. Can I use Full Production DNA and Voice DNA together?

Yes, when you have the rights and consent for both source sets. In the current primary workflow, Full Production DNA remains the home for catalog-based models and track references, while a trained Voice DNA model can be attached to that creator’s primary profile. Adding a Voice DNA does not replace the Full Production catalog already associated with that profile.

​

76. What is Model Based generation?

Model Based generation uses an available DNA model as the broader creative starting point. Choose it when you want the generation experience to draw from the overall musical identity represented by the selected DNA model. The actual controls still depend on what the selected DNA and model support in the product.

​

77. What is Track Based generation?

Track Based generation uses a specific processed reference from the DNA, such as a song, vocal, or instrumental reference. It is useful when you want a more focused starting point than the broader DNA model. The reference type does not guarantee that every output or control will be available. Use the options enabled for that reference.

​

78. Which should I choose: Model Based or Track Based?

Choose Model Based when you want the overall DNA model to guide the result. Choose Track Based when one available song, vocal, or instrumental reference is closer to the specific direction you want. Neither mode changes the ownership of the DNA, the rights you need for your prompt, or the payment rules for a later paid export.

​

79. Does generating with a DNA automatically create a sale or earning?

No. Creating or generating audio does not by itself create a creator earning. A creator earning is recorded only after a paid DNA export or a Fan Support payment is successfully verified through the product’s payment flow.

​

Earnings and payouts

​

80. What payment activity creates a creator earning in Earn?

The current Earn ledger records two sources:

• Paid DNA export
• Fan Support

Training a DNA, making it Public, getting Approved, generating audio, or opening a checkout page does not by itself add money to the ledger.

​

81. How does a paid DNA export create an earning?

When another creator completes payment to export audio made with an eligible paid DNA, the payment is verified and the DNA creator’s share is recorded in the protected earnings ledger. The history row is labeled Paid DNA export. Closing checkout, a failed payment, or an unpaid session does not create an export earning.

​

82. How much of a paid DNA export is reserved for the DNA creator?

The current policy reserves 50% of the gross purchase payment for the DNA creator. For example, a verified $10.00 paid DNA export reserves$5.00 for the creator before any later refund, dispute, or reversal adjustment. The amount is recorded as a protected creator earning, not an immediate bank transfer.

​

83. What is Fan Support?

Fan Support lets one person send a one-time or monthly Stripe support payment to a creator from an eligible Soundverse conversation. It is a support payment, not a DNA export purchase, ownership transfer, or commercial license. The recipient’s share is recorded in the protected creator-earnings ledger after the payment is verified.

​

84. How much of Fan Support is reserved for the creator?

The current policy reserves 75% of the gross verified support payment for the recipient. For example, a verified $5.00 one-time support payment reserves$3.75 for the creator before any later refund, dispute, or reversal adjustment. Earn shows the source as Fan Support.

​

85. Can a fan support a creator who has not connected Stripe yet?

Yes, when the conversation is eligible and the payment flow is available. The platform can record the recipient’s protected creator earning before that recipient has completed Stripe Connect onboarding. The recipient must still connect and complete Stripe setup before a payout can be sent to them.

​

86. Where does a pre-Stripe creator see a payment?

The creator can see the recorded amount in Earn history when the protected ledger is available. The row can carry a note such as Pending Stripe connection or Payout setup required. That means the earning is recorded but cannot yet be paid out to a Stripe account.

​

87. Does connecting Stripe later send all money immediately?

No. Connecting Stripe is an important payout requirement, but it does not bypass payment protection, the minimum payout threshold, account verification, or the payout schedule. Use the current Earn status and next-payout information for the account rather than expecting an instant transfer.

​

88. Can I send Fan Support to myself?

No. The payment flow rejects self-support. A supporter and recipient must be different people. This protects the support feature from being used as a self-payment path.

​

89. What one-time and monthly Fan Support amounts are available?

The current one-time presets are $2,$5, $10, and$25 USD. The current monthly presets are $3,$8, and $20 USD. The product can change available amounts over time, so use the options displayed in the current support dialog.

​

90. What is the difference between one-time and monthly Fan Support?

One-time support is a single Stripe payment. Monthly support uses a Stripe subscription and creates a recurring pledge until it is canceled or changes status. Each successful monthly payment is handled as its own verified payment for the recipient’s protected creator earnings.

​

91. Why does the product prevent a second monthly pledge?

The current flow prevents duplicate active monthly support from the same supporter to the same creator. If you already have an active pledge, manage that existing support instead of opening a second subscription. This avoids accidental duplicate recurring charges.

​

92. When does a paid export or Fan Support payment appear in Earn?

It appears after the payment provider reports a verified successful payment and the protected ledger records the creator’s share. Payment confirmation in a browser is not the only check. The payment must also pass the product’s provider verification and ledger recording steps. There can be a short delay while those systems complete their work.

​

93. Why did a fan pay but the creator does not see an earning yet?

First confirm that the checkout completed successfully and was not canceled or left unpaid. Then check the creator’s Earn history, the Stripe mode, and whether the protected earnings ledger is available. If it is still missing, record the approximate time, amount, visible status, and non-secret checkout or payment reference shown by the product for support. Never send card details, cookies, passwords, or Stripe credentials to support.

​

94. What does Current creator balance mean?

Current creator balance is the amount currently recorded in the protected creator-earnings ledger after known adjustments. It is not the available balance of a personal bank account and it is not a promise that the same amount can be withdrawn immediately. Some or all of it can still be in payment protection, awaiting Stripe setup, or awaiting a scheduled payout.

​

95. What is the difference between Money accrued and Money paid?

Money accrued is the running total of creator earnings that have been recorded in the ledger. Money paid is the running total that has been recorded as paid out. Accrued money can include earnings still in payment protection or awaiting a future payout. Paid money does not mean every current balance has been paid.

​

96. What does In payment protection mean?

The payment was verified and the creator’s share was recorded, but it is still being held during the applicable payment-protection period. The hold helps account for refunds, disputes, or other payment-provider events before the earning becomes payout-eligible. The length of that period can depend on the payment source and current product policy.

​

97. What does Eligible after protection mean?

The protected period for that balance has ended and the ledger considers the amount eligible for payout. It still needs a ready Stripe account, the required minimum balance, and an active payout path before money can move to the creator’s Stripe account. Eligible does not guarantee an immediate bank arrival.

​

98. What do Ready for payout, Payout processing, and Paid mean?

Ready for payout means the ledger considers the earning eligible after protection. Payout processing means a payout is currently being processed against the ledger balance. Paid means the ledger has recorded that payout as completed. Check Stripe as well as Earn for account-specific payout details and bank-arrival timing.

​

99. What does Payment review mean?

Payment review means the source payment is being checked after an event such as a refund or dispute. The creator amount can be adjusted or reversed while that review is resolved. It is not a confirmation that the payout will be sent unchanged.

​

100. What does Payout setup required mean?

The earning is recorded, but the payout account still needs an action. Common reasons include no connected Stripe account, incomplete onboarding, pending verification, restricted payments, missing bank details, or another Stripe requirement. Open the Stripe setup path from Earn and complete the requirement shown there.

​

101. What is the minimum payout threshold?

Earn displays the current minimum payout threshold for the account. The current product default is $50 USD, unless the connected account summary shows a different configured amount. Meeting the threshold alone is not enough: the balance must also be eligible after protection and the Stripe account must be ready for payouts.

​

102. When are automatic payouts scheduled?

When automatic payouts are enabled for a live account, the current product schedule is eligible balances at or above the displayed threshold on the 28th of each month. The Next payout date shown in Earn is the source of truth for the current account. Stripe and the receiving bank can add additional processing time after that scheduled date.

​

103. Why does Earn say Next payout date: Not scheduled?

The balance may be below the threshold, still in payment protection, awaiting Stripe setup, or the product may have automatic payouts disabled or no current payout schedule available. Review the payment status, threshold, and Stripe account state instead of assuming an unscheduled balance will transfer on the next calendar date.

​

104. What does it mean when automatic payouts are disabled?

Earn can continue recording protected and eligible creator balances while automatic payouts are disabled. It does not mean you should attempt to send yourself a transfer. It means the product is tracking the ledger balance while no automatic payout is being sent. Use the current Earn message and Stripe state for the account.

​

105. What is the difference between Stripe test mode and live mode?

Test or sandbox mode uses Stripe test accounts, test checkout objects, and test funds. It is for validating the flow and does not send real money to a bank. Live mode uses live Stripe accounts and live payment objects. Live earnings and payouts are separate from sandbox data. Do not mix a test checkout, test Connect account, or test payment method with a live flow.

​

106. Can I use a test Stripe account to verify Earn?

Use the Connect and Checkout flow started from the exact test or sandbox environment you are testing. That flow uses test-mode Stripe objects and test funds. Complete only the test onboarding information that Stripe requests for the test flow. Do not expect a test earning to appear as a live payout or to be withdrawable to a real bank account.

​

107. Why can a creator balance change after I first saw it?

The ledger can be updated when a payment is refunded, disputed, reversed, or otherwise adjusted by the payment provider. Earn can show a reversed amount or move an entry to Payment review. This is why a protected balance should not be treated as final cash before it reaches the appropriate payout status.

​

108. Does Public visibility or approval guarantee creator earnings?

No. Public visibility and Approved status determine whether a Full Production DNA can meet the current public-discovery condition. Earnings require a separate verified paid DNA export or Fan Support payment. Public discovery, a price field, a license choice, and Stripe connection do not by themselves create revenue.

​

109. What does Protected earnings are preparing mean?

It means the product cannot currently read the protected creator-earnings ledger. The page deliberately does not treat that condition as a zero balance, because zero would be misleading when the ledger is unavailable. Retry later. If it continues, note the time, account state, and visible error for support.

​

110. Where can I see the full earning history?

Open Earn and select View history. The history shows the earning source, date, status, and amount. It can include Paid DNA export, Fan Support, and recorded payouts to the Stripe account. Use the history status and any reversed amount to understand the current state of a specific earning.

​

111. What should I include when reporting an Earn or payout problem?

Include the account email, the page you used, whether it was test or live mode, the visible Earn status, approximate time and timezone, amount, source type, and any non-secret reference shown by Soundverse or Stripe. For Fan Support, also include whether it was one-time or monthly and whether the checkout completed. Do not include passwords, API keys, session cookies, full payment-card details, Stripe credentials, or identity documents.

​

112. Where can I learn the full earning and payout flow?

Use the Earn and Stripe Connect section in the DNA Help Center for the detailed lifecycle, status glossary, threshold, schedule, sandbox, refund, and Fan Support guidance. Use this FAQ when you need a short answer to one specific earnings or payout question.

​

Advanced Full Production DNA questions

​

113. How should I plan my first Full Production DNA catalog?

Start by deciding the musical identity you want the DNA to represent. Choose authorized songs, instrumentals, or acapellas that make sense together as a creator catalog or sound direction. Write down which files you intend to use, why each file belongs, and whether the material is ready to be shared with Soundverse. Planning before upload makes later review, troubleshooting, and follow-up uploads much easier.

​

114. Should every file in a Full Production catalog sound alike?

They do not need to be identical, but they should be intentional. A focused catalog gives the processing flow a clearer musical direction than a collection of unrelated files selected only to increase the file count. If you want to represent several distinct creative directions, decide whether they genuinely belong to the same creator identity before combining them. Use the processing result in My DNA to evaluate what the product actually created.

​

115. Can I upload more than one genre to Full Production DNA?

You can use authorized material from more than one genre when it represents the creator or catalog you want to model. However, broad genre mixing can make the intended direction harder to explain and the results harder to evaluate. Start with a cohesive direction when you are learning the flow, then add more material later if it remains appropriate for the same primary DNA. Do not assume that adding more genres guarantees more models or more public discovery.

​

116. Are more Full Production files always better?

No. The current flow can accept a large catalog, but quantity alone does not make a source collection more useful. Prioritize representative, authorized, and well-organized material over duplicate, unrelated, or low-value files. The processing outcome depends on the content of the catalog, not just the number of uploads.

​

117. Should I upload duplicate versions of the same song?

Avoid using duplicate files unless each version has a clear training purpose. For example, a final mix and an instrumental can represent different authorized source material, while repeated copies of the same exported master usually add little clarity. Keep a local source list so you can tell which version was included in a processing run. Do not rely on file name alone to prove what an upload contains.

​

118. Must Full Production source files be fully mastered releases?

No single mastering standard is promised by the flow. Use audio that clearly represents the sound, arrangement, and production identity you want the DNA to learn. Consistent, listenable source material is easier to evaluate than files with accidental silence, corruption, or unrelated content. The product does not guarantee a particular model result from a mastering choice.

​

119. Do Full Production uploads have to be complete songs?

No. The current Full Production flow accepts songs, instrumentals, and acapellas. Choose the form of audio that best represents the musical catalog you are authorized to use. An incomplete draft should be included only when it genuinely represents the intended creative direction.

​

120. Can I use instrumentals in Full Production DNA?

Yes. Instrumentals are an accepted Full Production source category and can help represent the production, arrangement, and musical texture of a catalog. An instrumental is different from a Voice DNA source recording, which is intended for one person’s clean solo voice. Use the flow that matches what you want the training to represent.

​

121. Can I use acapellas in Full Production DNA?

Yes. An acapella can be part of a Full Production catalog when it is authorized and belongs to the musical identity you want to represent. That does not automatically make it a Voice DNA dataset or guarantee a standalone vocal generation output. Use Voice DNA only for clean solo-vocal recordings when the goal is to train one person’s voice model.

​

122. Should I treat a stem as a Voice DNA recording?

Not automatically. A stem can contain backing sounds, multiple voices, effects, or material that does not meet the goal of clean solo-vocal voice training. Use a stem in Full Production DNA when it represents catalog material, or use a clean solo-vocal recording for Voice DNA when you have the required consent. Choose based on the training goal, not only the file label.

​

123. Can I use live recordings in Full Production DNA?

Use live recordings only when you have the authority to upload them and they represent the intended creator catalog. Audience noise, venue acoustics, and performance variation can make a live recording different from a studio production. Keep that distinction in mind when deciding whether the file belongs beside studio tracks in the same focused catalog. The product does not guarantee that a live recording will create a particular model or reference.

​

124. Can I upload audio that I found online?

Only if you own it or are authorized to use it for the selected DNA flow. Public availability on a website, social platform, streaming service, or video platform is not proof that you have training or upload rights. For Full Production DNA, confirm ownership and authority. For Voice DNA, confirm that the voice is yours or that you have the speaker’s consent. If rights are unclear, do not upload the material.

​

125. Can I include sampled or licensed audio in a DNA catalog?

Only when your agreement gives you the authority required for this use. Having permission to listen to, distribute, or commercially release a track does not automatically mean you have permission to use every component for DNA training. Review your contracts, sample licenses, and collaborator agreements before upload. This FAQ does not provide legal advice or expand any rights you may have.

​

126. Can I upload a collaboration to Full Production DNA?

Only if you have the necessary authority from all relevant rights holders or your agreement clearly permits the use. A collaboration can involve separate rights in composition, recording, vocals, production, samples, and artwork. Do not assume that being credited on a release is the same as having authority to use it for training. Keep any relevant permissions available in case you need to explain the source material later.

​

127. Why should I keep an inventory of my upload sources?

An inventory helps you remember what was included in a processing run. It can list a file name, original source, version, ownership or consent status, and the reason the file belongs in the catalog. That information makes it easier to decide what to add later and to prepare a clear support report if something fails. Do not upload private contracts or sensitive personal documents unless Soundverse provides a secure, specific request for them.

​

128. What should I check before selecting files?

Confirm the files are accepted audio formats, that you can identify them, and that you have the authority to upload them. Check that the collection matches the intended Full Production or Voice DNA goal. Keep local copies until upload and processing status are clear. Read the current product labels and confirmations before submitting the request.

​

129. How should I name source files before upload?

Use names that help you recognize the material later, such as creator, project, version, and date. Clear names are useful for your own source inventory and for explaining an issue to support. The file name does not replace the actual audio content, ownership confirmation, or product metadata. Avoid using passwords, account information, or other sensitive data in a file name.

​

130. Should I convert unsupported files before upload?

Use one of the accepted audio formats displayed by the current upload control. If a file is not accepted, make a lawful, high-quality conversion from a source you are authorized to use, then keep the original file available. Verify that the converted audio plays normally before attempting the upload again. Do not try to upload images, archives, documents, or video files through an audio input.

​

131. What should I do if an audio file does not play before upload?

Do not use it as a source until you understand the problem. Check the local file, file extension, and audio playback in a trusted player. If necessary, use a known-good authorized source or a properly converted copy in an accepted format. Include the visible product error and the file format in a support report if the issue continues.

​

132. What does upload progress confirm?

Upload progress confirms that the product is transferring the selected file data. It does not by itself confirm that processing has started, that a DNA result is ready, or that a public review is approved. Wait for the product to report that the upload is ready before starting Full Production processing. Use the next visible status in the flow rather than assuming an upload percentage is a training percentage.

​

133. What should I do if an upload is interrupted?

Keep the same source files available and retry through the product. The current Full Production uploader can use upload sessions and confirmed chunks when the session remains valid. If it cannot resume, the product can start a new session. Do not create duplicate or random files to work around the interruption. Record the visible error and approximate time if retries keep failing.

​

134. When is a Full Production upload ready to process?

Use the readiness state shown in the product. Processing should begin only after the uploaded audio has completed the required transfer and preparation state. Do not interpret a selected file name or partial upload progress as permission to start processing. If the Process DNA action remains unavailable, check the visible upload status and error text first.

​

135. Can I leave the page after a Full Production upload starts?

Follow the current product state rather than relying only on the browser window. Once a supported upload or processing operation is running, return later to My DNA or the relevant flow to inspect the result. Keep your source files until the flow reports a clear outcome. Do not close the page solely because you expect a particular result to be ready at a particular time.

​

136. Why can I not start Full Production processing twice?

The current flow blocks duplicate starts while a processing run is already active. This protects the same primary DNA from repeated accidental requests for the same work. Wait for the active state to complete or report a failure before deciding whether another run is needed. Starting multiple browser tabs does not create a valid shortcut around an active processing state.

​

137. Should I immediately run processing again after it starts?

No. Wait for the current processing state and inspect the resulting primary DNA before deciding whether more authorized material should be added later. A second run is useful only when there is a clear reason, such as new prepared catalog material, and no previous run is active. Do not use repeated starts as a substitute for reviewing the source collection.

​

138. Does a later Full Production run delete earlier results?

The current primary workflow is designed to add later uploads and processing results to the existing primary DNA. Earlier models and track results are intended to stay in place while new material is added. Inspect My DNA after the actual run to understand the current result for your account. Do not assume that every earlier result will be reproduced or that every new run will create a new visible model.

​

139. Will every processing run create a new DNA model?

No. The processing outcome depends on the supplied catalog and the result of the current run. The product can retain a usable primary model path even when it does not create several new style models or references. Check My DNA for the actual models and tracks that are available.

​

140. What determines the Full Production result?

The result depends on the authorized audio you supply, the processing outcome, and the current product capabilities for that DNA. Catalog quality, coherence, and the type of available source material can affect what is practical to review afterward. No documentation can promise a fixed number of models, tracks, tags, or generation controls before processing finishes. Treat the persisted My DNA result as the product source of truth.

​

141. What does the Full Production sensitivity control do?

Sensitivity is the catalog-processing control used in the Full Production flow. The current UI exposes a range from 0 to 10 and presents a balanced default. Use it as a configuration choice for the current processing request, not as a guarantee of better, faster, or more commercially valuable results. Read the guidance in the product before changing it.

​

142. When should I use a lower Full Production sensitivity?

Choose a lower setting only when it matches the intent described by the current product guidance and you understand why you are changing it. Start from the balanced value when you do not have a source-specific reason to choose a lower setting. Keep a note of the selected setting so you can evaluate the result later. Sensitivity is not a replacement for preparing a focused source catalog.

​

143. When should I use the default Full Production sensitivity?

Use the balanced default as a sensible starting point when you are testing a new catalog or do not have a specific reason to adjust the control. It provides a consistent baseline for comparing a later, intentional processing decision. The default does not guarantee a particular number of outputs or a particular public-review outcome. Review the finished result before deciding whether another authorized run is appropriate.

​

144. When should I use a higher Full Production sensitivity?

Use a higher setting only when it matches your documented training intent and the current product guidance. Do not change it simply because a higher number sounds universally better. Record the setting alongside the source catalog so you can make a meaningful comparison after processing. The actual outcome remains dependent on the catalog and available product behavior.

​

145. Does sensitivity guarantee a better DNA?

No. Sensitivity is one available processing setting; it cannot compensate for unclear rights, unrelated audio, incomplete uploads, or a source catalog that does not represent the intended sound. Use it together with thoughtful source selection and the visible product guidance. Evaluate the result in My DNA instead of relying on the setting name alone.

​

146. Why does the Full Production profile name matter?

A clear profile name helps you recognize the primary DNA later in My DNA, Discover, review surfaces, and support reports. Choose a name that is accurate, recognizable, and appropriate for the rights you hold. The profile name does not prove ownership, grant public approval, or guarantee discovery. Avoid using another person’s name or protected identity in a misleading way.

​

147. What should I write in a Full Production description?

Use the description to identify the creator, intended sound, or catalog context in a short and accurate way. The current form limits this field, so concise factual language works better than a long promise about outputs or earnings. Do not include private contact details, financial information, or claims you cannot support. The description helps recognition; it does not control the processing result by itself.

​

148. What do tags help with in a Full Production profile?

Tags help organize and describe the DNA in the product experience. Use accurate genre, mood, instrument, language, or creative-context terms that help people understand the intended profile. Tags do not guarantee search placement, public approval, generation quality, or creator earnings. Review and update them only when the current product allows it and the change remains truthful.

​

149. Do I need cover art for Full Production DNA?

Use cover art when it helps people recognize the profile in My DNA or public-facing product surfaces. The cover art is not the audio source used for Full Production processing and it does not replace the catalog upload. Only use art you own or are authorized to use. A cover image does not guarantee public visibility, approval, or an earnings outcome.

​

150. Can I change Full Production profile details later?

Use the controls available in the current product to review or update editable profile fields. Before changing a name, description, tag, visibility, price, or license selection, consider how it affects clarity for collaborators and future viewers. Changing metadata does not rewrite the source audio already used in a completed processing run. Read the current UI and applicable terms before relying on a metadata change for a commercial decision.

​

151. How should I choose initial visibility for a Full Production DNA?

Use Private or Personal while you are setting up, testing, or keeping the DNA outside the public-discovery path. Use Public only when you are ready for the DNA to be considered for public availability and you have met the necessary rights and product requirements. Visibility is separate from moderation status. Public does not automatically mean Discover, a sale, or a payout.

​

152. What does track visibility control?

Track visibility controls whether relevant uploaded or processed track references are visible in the Track Based product experience when that setting is available. It is a presentation choice, not a guarantee that every track creates a selectable reference or that every generation option will be enabled. Choose a setting that matches your intended sharing and privacy approach. Review actual visible results after processing.

​

153. What does Become Partner mean in the current Full Production flow?

Become Partner is a current configuration choice presented in the Advanced Options area. Use the exact product wording and applicable terms to understand its current effect before enabling it. It is not a standalone guarantee of earnings, public discovery, payout eligibility, or a specific revenue amount. Do not make business promises from a label alone.

​

154. What are Review Notes for?

Review Notes are an optional field in the Full Production review area. Use them to provide concise context that is relevant to reviewing the DNA, such as an intended catalog direction or clarification about the submitted material. Do not paste passwords, API keys, payment details, or private legal documents into Review Notes. Notes do not guarantee approval or a particular review duration.

​

Advanced Voice DNA questions

​

155. How should I plan a Voice DNA dataset?

Start with one specific voice and one clear goal: train a voice-focused model from authorized solo-vocal recordings. Gather clean recordings that represent the speaker or singer consistently, then confirm that you are the voice owner or have the speaker’s consent. Keep a local list of the source recordings, recording context, and consent status. Do not treat a mixed music catalog as a substitute for a voice-training dataset.

​

156. What does clean solo-vocal audio mean for Voice DNA?

It means audio where one intended speaker or singer can be heard clearly without competing voices, heavy background music, or distracting effects. Clean source material makes the training goal easier to understand and evaluate. The current guide recommends using one speaker or singer and aiming for about 10 minutes of usable voice audio. It does not guarantee a particular model result from any individual recording.

​

157. Should I remove background music from a Voice DNA recording?

Use recordings with little or no background music when the goal is a voice-focused model. Background music can make a source less clearly focused on the one voice you intend to train. If the musical context is itself the goal, consider whether the material belongs in Full Production DNA instead. Use the appropriate flow rather than trying to make one source type do both jobs.

​

158. Can I include two singers in one Voice DNA dataset?

No. Use one speaker or singer for one Voice DNA training request. Mixing multiple people makes it unclear which voice is intended to be trained and conflicts with the consent-focused purpose of the flow. Create separate authorized training plans for separate voices when the current product allows it.

​

159. Can one Voice DNA include several languages?

The source recordings can reflect the same authorized speaker or singer across relevant languages, accents, or vocal contexts. Keep the dataset focused on that one person and avoid treating language variety as permission to mix different voices. The current product does not promise a particular language, accent, pronunciation, or singing capability from a trained model. Inspect the available result and generation controls after training.

​

160. Should I include speaking and singing in the same Voice DNA dataset?

Choose source material that matches the voice behavior you intend to represent and follow the current product guidance. Mixed speaking and singing material can have different characteristics, so keep notes about what you supplied and why. Do not assume that any collection automatically produces every possible speaking or singing use case. Use the completed product result as the source of truth for what is available.

​

161. Can I use recordings from different microphones or sessions?

Use authorized recordings that clearly represent the same voice. Different microphones, rooms, and sessions can create variation, so keep the material as clear and consistent as possible when preparing the dataset. The product does not promise that every recording setup will produce the same result. Documenting your source sessions can help you evaluate a later training outcome.

​

162. Can I use a full song as a Voice DNA source?

Only when the audio is a clean, authorized solo-vocal recording that matches the Voice DNA goal. Most full songs contain production, instruments, backing vocals, or other material that may make them better suited to Full Production DNA. Do not label any vocal-containing song as a Voice DNA source without checking the actual audio. The distinction is based on the training goal and source content, not the file title.

​

163. Is an acapella always suitable for Voice DNA?

Not always. An acapella can still contain layered vocals, effects, ad-libs, crowd sound, processing, or more than one performer. Use it for Voice DNA only when it is a clean solo-vocal recording of one authorized voice and fits the current flow’s consent requirements. Otherwise, treat it as catalog material for Full Production DNA when appropriate.

​

164. How much Voice DNA audio should I prepare?

The current guidance is to aim for at least 10 minutes of clean solo-vocal audio. The form can accept at least one recording, but meeting the form minimum and building a useful voice dataset are different things. Use representative audio rather than assuming a very short or random clip will meet your creative goal. The guide does not promise a specific result from a particular duration.

​

165. Should I use one long Voice DNA recording or several files?

Use the source arrangement that best preserves clear, authorized audio of the same voice. Several recordings can help represent different clean sessions, while one recording can meet the basic form requirement. The trainer can re-slice recordings, so you do not need to manually cut every source into tiny clips before upload. Keep the original source files so you can retry or review them later.

​

166. Is one Voice DNA recording enough to train a useful model?

One recording can satisfy the basic upload requirement, but it may not represent enough clean, useful voice material for your creative goal. Follow the current guidance about using one speaker or singer and aiming for approximately 10 minutes of clean audio. Quality, clarity, consent, and relevance matter alongside duration. Treat the completed model and available controls as the real outcome.

​

167. Can I record Voice DNA audio directly in Soundverse?

The current Voice DNA upload area includes a Record control as well as file selection and drag-and-drop. Use a quiet environment and confirm that the recording contains only the authorized voice you intend to train. Review the recording before submitting it and keep any original source material you need for your records. Do not record someone else without their knowledge and consent.

​

168. Should I upload private personal recordings to Voice DNA?

Only upload material you are authorized to use and comfortable submitting through the current product flow. Avoid recordings that contain unrelated people’s voices, sensitive conversations, private addresses, passwords, financial data, or confidential information. If privacy is a concern, prepare a clean authorized recording specifically for the voice-training purpose. The product guide does not override your privacy, consent, or contractual obligations.

​

169. What does Voice DNA consent require?

The current Voice DNA flow requires confirmation that the voice is yours or that you have the singer’s consent to clone it. The checkbox is a statement you must be able to make truthfully. It is not a shortcut around obtaining permission, and it is not satisfied merely because a recording is publicly available. Do not begin training when consent is unclear.

​

170. Does consent for one recording always cover later Voice DNA uploads?

Keep consent and authority clear for each source set you use. If the source material, performer, agreement, purpose, or sharing context changes, review whether you still have the authority required for the new training action. Do not assume a previous confirmation resolves unrelated rights questions. Seek appropriate professional advice for legal or contractual questions.

​

171. Can I use a recording that contains background voices?

Avoid it for Voice DNA when the background voices are noticeable or could make the intended voice unclear. Voice DNA is designed around one speaker or singer, not a crowd, conversation, choir, or multi-performer recording. Use a clean alternate take when available. If you cannot identify the intended source voice cleanly, do not submit it as a Voice DNA dataset.

​

172. Does Voice DNA give me rights to another person’s identity?

No. Voice DNA is a product feature; it does not create ownership of another person’s voice, name, likeness, recording, or publicity rights. You remain responsible for having the required consent and permissions for the material and the intended use. Do not present a trained voice model in a misleading or deceptive way.

​

173. Why does the Voice DNA flow show a profile step?

The profile step provides the context for the voice-focused training request. It lets you identify the voice model through creator name, description, tags, cover image, and avatar choices available in the current product. Use accurate information that helps you recognize the result later. The profile does not replace the underlying consent or source-audio requirements.

​

174. How should I name a Voice DNA profile?

Use a clear, accurate name that helps you identify the voice model later in My DNA. Avoid names that falsely imply affiliation with another person, artist, or brand. The current flow can derive an internal voice-model name from the profile context, so you do not need to invent misleading numbered labels. Use the visible product name and status when reporting an issue.

​

175. What should I write in a Voice DNA description?

Write short, accurate context that helps you recognize the authorized voice model later. Avoid unsupported promises about resemblance, commercial rights, language coverage, output quality, or earnings. Do not include private contact information, sensitive personal information, or someone else’s identity details without a proper reason and authority. The description helps organization; it does not define the training outcome by itself.

​

176. Does Voice DNA cover art affect the trained voice?

No. Cover art and avatar choices help identify the profile visually, but the voice-training source is the authorized audio recording. Only use images you own or are authorized to display. Do not use someone else’s likeness or branding in a way that is misleading or unauthorized.

​

177. What are Voice DNA settings for?

The Settings step exposes the current voice-profile options available in the product, including visibility choices where present. Use the current labels rather than assuming that Voice DNA has exactly the same controls as Full Production DNA. Voice DNA currently presents private and public visibility options in its form. Visibility does not replace consent, rights, moderation, or product availability requirements.

​

178. What is Voice DNA quality?

Voice DNA quality is the epoch-based training control used for the voice model request. The current range is 10 through 350 epochs, with a balanced default around 180 epochs. It is different from the Full Production sensitivity control. Use the product guidance and your prepared source material when choosing a setting.

​

179. What does the Fast Voice DNA quality label mean?

Fast corresponds to the low end of the current epoch range, beginning at 10 epochs. It can be useful for a quicker, lower-epoch training choice when that fits the current product guidance. It does not promise a finished result by a fixed time or guarantee that a smaller setting is right for every dataset. Record the selected value if you want to compare training decisions later.

​

180. What does the Balanced Voice DNA quality label mean?

Balanced is the current middle guidance, at about 180 epochs. It is a sensible starting point when you do not have a source-specific reason to choose a lower or higher setting. Balanced does not guarantee a particular voice resemblance, language capability, duration, or commercial use outcome. Use the completed result and current product controls to evaluate it.

​

181. What does the Studio Voice DNA quality label mean?

Studio corresponds to the high end of the current range, at 350 epochs. The UI describes more epochs as seeking a higher-fidelity clone while taking longer to train. Higher is not automatically better for every source set, and it does not overcome unclear or noisy source audio. Choose it intentionally and review the available cost and duration estimate before starting.

​

182. Are Voice DNA duration and token figures guarantees?

No. The DNA Quality panel can show estimated duration and token cost for the chosen epoch value. Those are current UI estimates, not guarantees of capacity, final charge, training success, or exact completion time. Check the visible estimate immediately before starting the request.

​

183. What does Queued mean in Voice DNA?

Queued means the voice-training request is waiting for available capacity. The current UI describes this as waiting for a GPU to become available. Queued is not the same as failed, and it is not a reliable countdown to completion. You can return later to check the current training state.

​

184. What does Preparing mean in Voice DNA?

Preparing means the system is packaging or processing the voice dataset before reported epoch progress is available. The UI can show an indeterminate progress state during this phase. Do not treat that state as a fixed percentage or a promise of a completion time. Keep the source files and check the product status later.

​

185. What does Training mean in Voice DNA?

Training means the product has begun reporting the voice-training progression. When epochs are available, the UI can show current and total epoch information. That percentage relates to reported training progress; it is not an exact wall-clock completion forecast. Avoid creating duplicate requests while the current training run is active.

​

186. What does Completed mean in Voice DNA?

Completed means the current voice-training result reports completion. Use View in your profile or My DNA to inspect the actual trained model and the controls available for it. Completion does not independently promise public discovery, every output style, earnings, or commercial rights. Review the visible model state rather than relying only on a notification.

​

187. What does Failed mean in Voice DNA?

Failed means the current training request reported a failure. The product directs the flow back to DNA Quality so you can review the visible error, source recordings, consent, and selected quality before retrying. Do not submit repeated retries without understanding what changed. Include the visible error, approximate time, and non-sensitive source details if you contact support.

​

188. Can I close the Voice DNA window during training?

The current UI states that Voice DNA training continues when you close the window. Return later to the profile or My DNA to inspect the current state. Keep the original recordings until the outcome is clear. Do not rely solely on an email or browser tab remaining open as proof that the model is ready.

​

189. What should I do before retrying a failed Voice DNA request?

Read the exact status and visible error first. Check that the source files are accepted audio formats, that the dataset contains one authorized voice, and that the consent confirmation is truthful. Review the selected epoch value and any available token or duration estimate. Retry only after making an intentional correction or after a temporary product issue has cleared.

​

190. Why is Voice DNA progress not changing yet?

The flow can remain in Queued or Preparing before reported epoch progress begins. These are different stages from active Training and do not provide a guaranteed completion timer. Return later to check the current state rather than refreshing repeatedly or starting duplicate requests. If the status remains unclear, collect the visible state, approximate time, and non-sensitive details for support.

​

191. Can I create more than one Voice DNA model?

Use the options and profile context currently available to your account. The current primary workflow can attach trained voice models to the canonical primary DNA profile without replacing the Full Production catalog content. Do not assume that a new request creates an unrelated primary Full Production profile. Review My DNA to see the actual available voice-model results.

​

192. Can I change a trained Voice DNA model after completion?

Use the controls the current product makes available for the completed model and profile. Changing visible profile details does not retroactively change the source audio or the completed training process. If you need a different training result, prepare authorized source material and follow the current training flow rather than altering records outside the product. The product does not promise that every edit creates a new model.

​

193. How do I use a completed Voice DNA model?

Open My DNA or the relevant DNA generation experience and select the model only when the product presents it as available. Use the output styles and input fields enabled for the selected context. Do not assume that every Voice DNA model exposes every possible singing, speech, or standalone vocal option. The selected model and current product controls determine what you can request.

​

194. What is the difference between a Voice DNA model and a vocal track reference?

A Voice DNA model is trained through the dedicated voice-focused flow from clean recordings of one authorized voice. A vocal track reference is a processed reference that can appear as part of a DNA’s catalog context. A vocal reference does not automatically mean a standalone Voice DNA model exists, and a Voice DNA model does not guarantee every vocal reference behavior. Use the labels and controls shown in the current product.

​

195. What should I check when a Voice DNA result looks different from expected?

Review whether the source material was clear, authorized, focused on one voice, and aligned with the intended goal. Check the selected epoch value, visible training state, and the completed model shown in My DNA. Do not measure the result only against a single short clip or a promise that was not present in the product. Use a carefully prepared future source set only when you have a clear reason and authority to do so.

​

196. What should I send support for a Voice DNA issue?

Include the profile or model name, the step where the issue occurred, visible error text, approximate time and timezone, selected quality, training state, and audio formats or recording count. State whether the issue involved upload, consent, settings, training, completion, or generation. Do not send raw private voice recordings, passwords, session cookies, API keys, Stripe credentials, or identity documents through an ordinary support message. Use a secure channel only if support specifically asks for private material.

​

Advanced generation, discovery, and publishing questions

​

197. How do I choose a DNA for generation?

Choose a DNA that is available to your account and that matches the musical or voice context you intend to explore. Then review the models and track references that the product actually presents for that DNA. Do not assume that every DNA exposes the same selection list, output style, prompt field, or export option. The enabled controls are the reliable guide for the current selection.

​

198. How do I choose between an available DNA model and a track reference?

Choose a DNA model when you want the broader processed DNA identity to guide the generation. Choose a track reference when one available song, vocal, or instrumental reference is closer to the focused starting point you want. Neither selection guarantees a particular output, and changing the selection can change the controls shown. Use the result you actually hear and the product’s enabled options to evaluate the next step.

​

199. Why does changing a model change the generation controls?

Generation controls are context-aware. They can depend on the selected DNA, the available model, the chosen track reference, and the capabilities currently returned for that selection. If an option disappears or becomes disabled, it is not currently supported for that context. Choose another compatible available selection only when it fits your intended creative direction.

​

200. Why does changing a track reference change the generation controls?

A track reference is a more specific processed starting point than a broader DNA model. Song, vocal, and instrumental references can have different available contexts in the generation experience. The UI does not promise that every reference enables every output style or input field. Use the controls visible after selecting the reference instead of assuming a generic feature matrix.

​

201. What should I put in a DNA generation prompt?

Describe the musical idea you want to create using clear creative details that fit the enabled generation flow. You can describe mood, arrangement, pacing, instrumentation, energy, lyrical idea, or other relevant production intent when the product asks for it. The prompt does not change your rights in the selected DNA, and it does not override restrictions or capabilities of the selected context. Avoid including private information or instructions that you do not have the right to use.

​

202. Should I use lyrics when the product shows a lyrics field?

Use lyrics only when the selected DNA generation flow presents a lyrics field and the content matches your intended request. Make sure you have the right to use the lyric content you submit. Lyrics can guide the current request, but they do not guarantee a specific vocal output, melody, language behavior, or commercial result. Follow the visible product controls for the selected DNA and output style.

​

203. Can I paste someone else’s lyrics or prompt into DNA generation?

Only use material you own, have permission to use, or are otherwise legally entitled to submit. Public availability is not the same as permission to reuse a lyric, prompt, script, recording, or artist identity. Do not use the product to impersonate people or make deceptive claims about authorship or collaboration. This documentation does not provide legal advice about a specific work.

​

204. What does the Song generation style mean?

Song is one of the current generation styles supported by the DNA request contract. It refers to the available song-generation path for the selected DNA context. Whether lyrics, a specific reference type, or other controls are available still depends on the current selection. Use the enabled Song controls rather than expecting every DNA to behave the same way.

​

205. What does the Instrumental generation style mean?

Instrumental is the other current generation style supported by the DNA request contract. It refers to the available instrumental-generation path for the selected DNA context. The presence of an instrumental reference does not guarantee every model, prompt, or export option will be available. Use the actual selection and enabled controls in the product.

​

206. Why is there no separate universal Vocal generation style?

The current DNA generation request contract supports Song and Instrumental styles. The interface can show vocal references, but a vocal reference is not the same thing as a universal standalone vocal or acapella request style. Do not promise a specific vocal-only output merely because a vocal reference appears in the selection list. Use the output styles enabled for that exact context.

​

207. Can a vocal reference guarantee a standalone vocal output?

No. A vocal reference can be useful context within the current DNA experience, but it does not guarantee that the backend supports a separate standalone vocal output for every selection. The selected DNA, model, reference, and enabled controls determine what can be requested. Check the product rather than relying on a generic label.

​

208. Why is an output option disabled?

An output option can be disabled because the selected DNA, model, or reference does not currently support it. It can also reflect current product availability or an incomplete processing result. Review My DNA and switch only to another available selection that genuinely matches your goal. Do not treat a disabled option as proof that the DNA has failed or that a payment is required.

​

209. How can I compare different DNA generation choices fairly?

Keep one part of the request consistent while changing one selection at a time, such as model versus track reference. Record the selected DNA, model or reference, output style, and prompt so you can understand why two results differ. Listen to the actual outputs instead of judging only from the name of a setting. Do not assume that a comparison can prove a universal quality ranking for all DNA content.

​

210. Can I use my own DNA for generation?

Use your own available DNA through the same product controls that apply to the current selection. Your ability to choose a DNA does not change the need to respect the applicable source, profile, licensing, and platform rules. The available models, references, and generation styles are determined by the actual processing result. Check My DNA when you need to confirm the current available assets.

​

211. Can I use another creator’s DNA?

Use another creator’s DNA only through the options that Soundverse makes available to your account. Respect the displayed price, visibility, license, export, and product conditions for that DNA. Do not claim that another creator endorsed, collaborated on, or owns your newly generated work unless that is actually true. The product’s available selection does not override applicable terms or rights requirements.

​

212. Does selecting a DNA make my output identical to the source material?

No such guarantee is made. DNA provides available creative context through processed models or track references, while the actual generated result depends on the selected flow and current product behavior. Do not present a generated result as an original recording by a real artist or as a copy of a particular source track. Use honest labels and credits appropriate to the work and applicable terms.

​

213. What is Discover for?

Discover is the Soundverse area for browsing DNA content that is eligible for public discovery. It is separate from My DNA, which is the main area for managing your own DNA work. Discover availability depends on the current product state, visibility, approval status, and the DNA content presented to your account. Do not use public browsing as proof of ownership or commercial rights.

​

214. Why is my Full Production DNA not in Discover?

For the current Full Production public path, the DNA must be both Public and Approved. Private, Personal, Pending, Rejected, and On hold states are not eligible for that public-discovery condition. Check the actual visibility and moderation status shown for the DNA. This guide does not promise a review outcome or review duration.

​

215. Does making a DNA Public publish it immediately?

No. Public visibility is necessary for the current public-discovery path, but Full Production DNA must also reach Approved status. Public visibility does not guarantee Discover placement, a buyer, an export, a Fan Support payment, or a payout. Use the product’s displayed state before describing a DNA as publicly discoverable.

​

216. What is the difference between Public, Private, and Personal in Full Production DNA?

Public is the visibility choice used when the DNA is intended to be considered for public availability. Private and Personal keep the DNA outside the current public-discovery path and can be useful while setting up, testing, or keeping the DNA non-public. These are visibility choices, not moderation outcomes. Public discovery still requires the required Approved status.

​

217. Does a Private DNA avoid all product processing?

Visibility controls where the DNA is intended to appear in the product. It is not a statement that uploaded audio has not been processed or that no product records exist. Use Private or Personal when you do not want the DNA considered for the public-discovery path. Keep your source authorization and privacy decisions clear before upload. Use the product’s current labels and settings for the exact account.

​

218. What does Pending mean for my DNA?

Pending means the Full Production DNA has not reached the Approved state required for current public discovery. It is different from a processing progress indicator and different from a Stripe or payout status. Do not promise a public launch date while the product shows Pending. Review the current status rather than relying on a previous screen or notification.

​

219. What should I do if a DNA is Rejected or On hold?

Read the current product status and any visible explanation before taking another action. Do not assume that changing the visibility setting alone will bypass a Rejected or On hold status. Review source rights, profile details, and the relevant guidance; contact support with the visible status and non-sensitive details if necessary. Do not upload substitute material without first understanding the issue.

​

220. Does an Approved DNA guarantee earnings?

No. Approval and Public visibility can make a Full Production DNA eligible for the current discovery path, but earnings require a separate verified paid DNA export or Fan Support payment. Price, license, Partner selection, and Stripe connection also do not guarantee a buyer, revenue amount, or payout date. Use Earn history for recorded activity rather than treating approval as income.

​

221. Does a DNA price guarantee what a buyer receives?

The price control is part of the current product configuration and can affect what is presented in the product. Read the current pricing, license, and export details before relying on them for a commercial decision. The field does not independently define every legal, ownership, payout, or entitlement outcome. Use the product terms and completed checkout or export state for the actual transaction result.

​

222. Does a DNA license selection guarantee commercial rights?

No. The current Full Production form presents license labels, but a UI selection does not independently explain every commercial or legal effect. Read the exact current license information and applicable Soundverse terms before releasing, distributing, or monetizing work. Seek qualified legal advice for questions about a particular agreement or release.

​

223. Can I share a DNA with collaborators before it is public?

Use only the sharing and visibility options the current product makes available to you. Do not share source audio, a private profile, or someone else’s voice without the authority to do so. Private or Personal visibility does not create an automatic private-license arrangement with collaborators. Agree on permissions and use the appropriate product controls for the actual relationship.

​

224. What should I tell collaborators about a DNA’s status?

Use the actual product state: processing, Pending, Approved, Private, Personal, Public, Queued, Training, or the current Earn status. Avoid promising a review date, public listing, exact generated output, sale, or payout before the product reports it. Clear status language prevents confusion and helps collaborators make informed decisions. Do not share private account details or Stripe information in ordinary project chats.

​

225. Can I use a DNA name or cover art that resembles another artist?

Avoid names, images, descriptions, or branding that could mislead people about affiliation, ownership, or endorsement. Use only names and artwork you own or are authorized to use. The product profile should accurately identify the creator and intended sound direction. This applies even if the related material is not currently public.

​

226. What should I do if My DNA does not show the result I expected?

Check the source upload state, processing or training status, and whether you are looking at the primary DNA, a model, a track reference, or a Voice DNA result. For Full Production, remember that later uploads update the existing primary profile rather than creating a second primary profile. For Voice DNA, check whether the state is Queued, Preparing, Training, Completed, or Failed. Capture the visible status and approximate time before contacting support.

​

227. Why is a generation result different from the description I wrote?

Generation is context-dependent and can vary with the selected DNA, model, reference, enabled output style, prompt, lyrics, and current product behavior. Use a clear prompt and test one intentional change at a time when comparing results. Do not promise an exact arrangement, voice behavior, or replication of a source recording from a text description alone. Review the actual result and the options enabled for the current selection.

​

228. What should I do before exporting generated DNA audio?

Review the current product details, price, license, and any checkout information shown for the selected audio. Make sure you understand whether the action is free, paid, or restricted for your current selection. Do not assume that a generation preview, a public DNA, or a connected Stripe account automatically grants an export entitlement. Use the completed product state and applicable terms for the specific export.

​

229. What should I do after a paid DNA export completes?

Use the completed product experience to confirm the buyer’s entitlement and the creator’s Earn history where applicable. The creator’s ledger entry can initially be in payment protection rather than immediately ready for payout. Keep non-sensitive transaction details available in case you need support, but do not share payment card information or credentials. Check the current export and Earn status rather than relying only on a browser redirect.

​

230. Why might a paid DNA export not create an immediate payout?

The creator share is first recorded in a protected earning lifecycle. It can remain in payment protection, wait for Stripe setup, need to meet the displayed payout threshold, or wait for the live payout schedule. Refunds and disputes can also affect the amount before payout. Use Earn history to distinguish a recorded creator earning from a completed Stripe payout.

​

231. How can I build a reliable DNA workflow?

Use a repeatable sequence: prepare authorized sources, keep an inventory, choose the correct DNA type, document selected settings, monitor the visible status, and review My DNA before making another request. For public sharing, separately verify visibility and approval. For money, separately verify Earn history, Stripe setup, protection status, threshold, and schedule. Keeping these stages separate prevents many avoidable misunderstandings.

​

232. What should I document for each DNA project?

Record the profile name, source-list summary, authorization or consent confirmation, selected settings, upload date, processing or training status, and the actual resulting models or references. For public DNA, note the visibility and moderation status. For paid activity, note the non-sensitive Earn source and status. This record is for your own organization; do not store passwords, Stripe secrets, or sensitive identity documents in it. Use a secure process for any sensitive creator records you must keep.

​

233. What expectations should I avoid setting with fans or collaborators?

Avoid promising exact training time, review time, model count, output quality, public discovery, buyer activity, royalty amount, payout date, or bank-arrival date. Avoid claiming someone else’s consent, authorship, identity, or endorsement without a real basis. Use the current product state and applicable terms when describing what has actually happened. Honest, specific status updates are more useful than broad promises.

​

234. How do I prepare a clear DNA support report?

State whether the issue is Full Production DNA, Voice DNA, generation, Discover, a paid export, Fan Support, Stripe Connect, or Earn. Include the affected profile or model name, exact page or step, visible error, approximate time and timezone, and relevant non-sensitive status. For uploads, include the number and formats of files. For money, include source type, amount, mode, and whether it was test or live. Do not include secrets, full payment-card data, session cookies, passwords, API keys, or raw private audio unless support provides a secure request.

​

235. Where should I look first when something changes in DNA?

Look at the relevant in-product source of truth: My DNA for models and training results, Discover for public availability, the generation view for current controls, and Earn for creator-payment history. Then check the actual status label rather than relying on a past email, browser cache, or assumption. If the product shows an error, save the non-sensitive text and the approximate time before retrying. This creates a useful starting point for support without exposing private account information.

​

236. Where can I find detailed DNA guidance beyond this FAQ?

Use the DNA Help Center for the end-to-end walkthrough, planning playbooks, glossary, lifecycle explanations, and troubleshooting paths. Return to this FAQ for quick answers about a specific choice, status, or visible product behavior. For account-specific problems, use the support checklist and provide only the non-sensitive information needed to investigate. Keep the current product screen and applicable terms as the source of truth for your exact account.