Guide scope
This guide explains the current Soundverse DNA creation and use flows. It covers:- Full Production DNA
- Voice DNA
- DNA models and track references
- Processing and training states
- Visibility and public discovery
- Earn and Stripe Connect boundaries
- Common troubleshooting steps
Quick decision
Use Full Production DNA when you want to train from a catalog of songs, instrumentals, or acapellas. Use Voice DNA when you want to train from clean solo-vocal recordings. Do not use Voice DNA as a shortcut for a mixed music catalog. Do not use Full Production DNA as a shortcut for cloning a particular person’s voice. Only upload material that you own or are authorized to use. For Voice DNA, you must also be the speaker or have the singer’s consent.
Product map
The DNA area contains several related surfaces. They have different purposes.My DNA
My DNA is the main management area for your own DNA work. Use it to inspect your primary DNA, models, tracks, and voice models. It is the place to return after a Full Production processing run or Voice DNA training run. Treat the current in-product state in My DNA as the source of truth for your result.Discover
Discover is for browsing DNA content that is eligible for public discovery. Public visibility alone does not make a DNA discoverable. The current Full Production public path requires both Public visibility and Approved status.Earn
Earn contains the current Stripe Connect flow where creator earnings are available. It can show a connection state and the DNA dashboard summary available for the connected account. It does not independently guarantee royalties, payout dates, eligibility, or commercial rights.Generation
The DNA generation experience uses an available DNA model or track reference. The controls change according to the selected DNA, model, and reference. Use the controls that the product enables for the current selection.Core concepts
DNA
DNA is a Soundverse creator profile built from audio. It provides source material that Soundverse can process into DNA models and track references. The number and type of resulting assets vary with the supplied material and processing outcome.Primary DNA
The current Full Production flow uses one primary DNA profile per owner. That profile is the ongoing home for the creator’s catalog-based DNA. Later uploads and processing runs add to the same primary DNA. They do not create another primary Full Production DNA profile for the same owner.Full Production DNA
Full Production DNA is the catalog-based flow. It accepts songs, instrumentals, and acapellas. It processes the catalog into available music-style models and track references. Use it when the goal is to represent a broader musical identity or catalog.Voice DNA
Voice DNA is the voice-focused flow. It accepts clean solo-vocal recordings and trains a voice model. In the current primary workflow, the trained voice model is added to the creator’s primary DNA profile. It does not remove or replace existing Full Production catalog material.DNA model
A DNA model is an available processed model associated with a DNA. When the product presents it, it can be selected as a broader starting point for generation. Model availability depends on processing and on the current DNA.Track reference
A track reference is an available processed audio reference associated with a DNA. The product can show song, vocal, or instrumental reference types. A track reference can narrow the starting point for available generation. It does not guarantee that every generation control is supported.Generation style
The current DNA generation contract supports two request styles:- Song
- Instrumental
Part 1: Prepare before creating DNA
1. Decide which flow you need
Start by deciding what you want the DNA to represent. Choose Full Production DNA for a music catalog. Choose Voice DNA for one person’s voice. Do not mix the two goals together without understanding which flow owns the training request.2. Confirm authorization
Before selecting any files, confirm that you have the necessary authority to upload them. For Full Production DNA, the current upload form asks you to confirm ownership and authority to upload the files. For Voice DNA, the form asks you to confirm that the voice is yours or that you have the singer’s consent. Do not submit a confirmation you cannot truthfully make.3. Keep the original source files
Keep a local copy of every selected audio file until the upload and processing state are clear. This makes retrying an upload easier if a transfer is interrupted. It also helps you identify exactly which material was included when you contact support.4. Prepare a focused Full Production catalog
Use songs, instrumentals, or acapellas that represent the sound you want the DNA to learn. The source collection does not need to be enormous to be useful. It should be intentional. A coherent catalog is generally easier to reason about than unrelated audio selected only to raise the file count.5. Prepare clean Voice DNA recordings
Aim for at least 10 minutes of clean solo-vocal audio. Use one speaker or singer only. Avoid background music and heavy effects. The voice trainer can re-slice recordings. You do not need to pre-cut every source recording into small clips before upload.6. Choose a clear name
Give the primary DNA or voice profile a name you can recognize later. The name should help distinguish the creator or voice model in My DNA. Avoid relying on a vague temporary name when you expect to create several models or return later for follow-up training.7. Prepare short profile copy
The current profile-description inputs are limited to 300 characters. Prepare a concise description that helps identify the DNA. Use tags to summarize the musical identity or profile context. Tags help organize the product experience but do not guarantee a specific generated result.8. Choose an initial visibility intentionally
Full Production DNA and Voice DNA do not present exactly the same visibility controls. For Full Production DNA, the current choices are Public, Private, and Personal. For Voice DNA, the current form presents private and public. Use a non-public setting while you are training or testing if you do not want the DNA considered for public visibility.Part 2: Full Production DNA
Overview
Full Production DNA is the primary catalog-based flow. It has five current steps:- Profile Settings
- Upload Section
- Process DNA
- Advanced
- Finalize
Step 1: Open the primary profile
Open the Full Production DNA flow from the DNA area. The app retrieves or creates the canonical primary DNA profile for the current owner. This is why the flow does not create a second primary profile for the same owner. Existing legacy DNA entries can still be visible for some accounts. The current Full Production creation path remains focused on the primary DNA.Step 1: Complete Profile Settings
The first Full Production panel is the profile area. Use it to identify the primary DNA before uploading more source material. The current fields include creator name, description, tags, and cover art.
Creator name
Enter a name that clearly identifies the profile. This is the name you will see again when managing the DNA. Use a stable creator or project name rather than an ambiguous file label.Description
The Full Production description is limited to 300 characters in the current form. Use it to explain the DNA in a compact way. It is better to describe the broad musical purpose than to promise a result that processing cannot guarantee.Tags
The profile area offers suggested tags and a tag input. Use tags to describe the catalog or creative direction. Examples of useful tag themes include genre, mood, instrumentation, era, or production character when they are actually relevant to the uploaded work. Tags are metadata. They are not a promise that a corresponding model or output style will be created.Cover art
The Full Production profile includes cover-art controls. Use a cover image when you want to make the primary DNA easier to recognize. The cover art does not replace the audio catalog. Full Production processing uses the uploaded music files rather than the cover image.Check the profile before continuing
Before moving to Upload Section, confirm:- The creator name is recognizable
- The description is concise and accurate
- The tags make sense for the catalog
- The cover image is appropriate if you chose one
Step 2: Upload Section
Use Upload Section to add audio for Full Production DNA processing. The current flow accepts at least one audio file. The current maximum is 1,000 audio files. The upload text does not require a minimum duration for Full Production audio.
Full Production accepted formats
The current Full Production file input accepts:- .mp3
- .wav
- .aac
- .ogg
- .flac
- .m4a
Full Production source categories
You can use:- Songs
- Instrumentals
- Acapellas
Ownership confirmation
The Full Production upload step includes an ownership-and-authority confirmation. Read it before continuing. Confirm it only when you have the authority to include every selected file. If even one file is unclear, remove it until you resolve the rights question.Choose a useful catalog
Choose files that represent the intended DNA sound. The product can process varied catalog material, but results can vary when source audio has no clear connection. Do not expect a guaranteed number of models or track references from a particular file count. The collection’s relevance matters more than a raw file total.Watch upload progress
The current Full Production uploader reports progress as it transfers the selected files. Wait for the upload flow to confirm that the selected audio is ready before trying to process it. Do not treat a file-selection action as proof that the file is fully uploaded.
Temporary interruption and retry
The current Full Production uploader uses an upload session and local checkpoints. When the session remains valid, it can reuse confirmed chunks instead of blindly restarting every file transfer. If a temporary network problem appears, retry with the same files. Keep the source files in place until the product confirms completion.When a session cannot resume
An upload session can expire or become invalid. In that case, the uploader can request a new server-issued session. This is expected recovery behavior. It does not mean that the previous session should be treated as a completed upload.When the upload reports an error
Read the error message before taking the next action. Check that the selected files are accepted audio formats. Keep the file list available. Retry the same selection once the product allows it. If the failure persists, capture the error text, time, and number of files for support.Do not process incomplete uploads
Processing requires audio that is ready to process. Do not click Process DNA as soon as files are selected if the uploader is still transferring them. Wait until the upload UI indicates that the source material has completed its upload path.Step 3: Process DNA
Process DNA is the step that starts Full Production model processing. It is not merely a review screen. The primary flow supplies the selected sensitivity and the newly uploaded audio to the processing request. Use this step only after the upload is ready.Full Production sensitivity
The current sensitivity slider ranges from 0 to 10. The default value is 5. The UI describes lower sensitivity as creating broader genre DNA results. It describes higher sensitivity as creating more niche style DNA results.
Choosing a lower sensitivity
Choose a lower sensitivity when the goal is broader genre-style grouping. This is a creative configuration choice. It does not create a separate primary DNA profile. It does not promise that a particular broad genre name will appear in the result.Choosing the default sensitivity
The current default is 5. The interface presents that central value as the balanced choice. It is a reasonable place to begin when you do not have a clear reason to prefer broader or more niche grouping.Choosing a higher sensitivity
Choose a higher sensitivity when you want the catalog processed toward more niche style DNA results. Higher sensitivity is not a guarantee that each source track becomes its own style model. The actual results still depend on the processed catalog.Start processing once
The current primary flow prevents a second processing start while DNA processing is already active for the same DNA. If the UI says processing is already running, wait for the active run to complete. Do not repeatedly launch duplicate attempts while the first run is still active.What happens to earlier results on a later run?
The current primary processing path is additive. It preserves existing core playlists rather than clearing them before re-clustering. Later uploads can add new models and track results to the primary DNA. The earlier content is intended to remain in place.What Process DNA does not promise
It does not promise:- A fixed completion time
- A fixed model count
- A fixed track count
- Every possible reference type
- A specific genre or tag
Step 4: Advanced
The Full Production Advanced panel has three current areas:- Advanced Options
- Tagging & Categorization
- Review
Advanced Options
The current Advanced Options form includes:- Visibility
- Price
- License
- Tracks
- Become Partner
Visibility in Full Production DNA
The current Full Production visibility values are:- Public
- Private
- Personal
Price and license
The current Full Production form includes a price input and a license selection. The current license labels shown are Full ownership, Standard, and Distribution. The legal effect of those values is governed by the current product terms and license terms. Do not publish, distribute, or make commercial promises based only on a label in this guide.Track visibility
The current form includes a Tracks setting with Visible and Hidden choices. Use the value shown in the product for the content you intend to expose. Do not assume that this field alone overrides every other visibility, moderation, or access rule.Become Partner
The current Advanced Options area includes a Become Partner checkbox. Use the product’s current details and terms to understand its commercial effect. This guide does not promise that selecting it creates earnings eligibility or a payout.Tagging and Categorization
The Tagging & Categorization panel includes:- Primary Category
- Subcategory
- Tags
Review
The current Full Production review area includes tabs for:- Models
- Tracks
- Voice Models
Review before moving on
Review the displayed data before completing the flow. Check that your visible configuration matches your intent. Review is useful for finding a wrong name, tag, category, price, visibility value, or source choice before you rely on the finished DNA. The Review panel is not a promise that the DNA has been approved for public discovery.Step 5: Finalize
Finalize is the final part of the Full Production creation interface. The primary flow starts catalog processing from Process DNA before Finalize. Use Finalize to complete the current configuration and review experience. Do not assume that Finalize restarts processing if processing has already begun.After Finalize
Return to My DNA to inspect the current result. The product can add model and track results to the primary DNA as processing completes. Results can appear as available models, tracks, or voice-related content in the management experience.Model-ready email
Soundverse attempts to send a model-ready email after a full-production model run has persisted results. Email is best effort. The model result is queryable in the product before an email delivery issue can affect it. If the email does not arrive, check My DNA rather than assuming the model failed.When you can leave the page
Once Full Production processing has started, you can leave the page and return later. The current product flow continues the processing work independently of keeping the creation view open. My DNA is the right place to inspect the outcome later.Part 3: What Full Production processing can create
Models
Full Production processing can add music-style models to the primary DNA. The models available for a DNA are driven by the processing result. One catalog can produce a different model result from another catalog. Do not promise a specific count before the processing result is available.Track references
Processing can associate track references with the primary DNA. The interface can represent available source references as song, vocal, or instrumental. Those are reference categories. They are not a universal set of final generation buttons.Tags and classifications
The processing pipeline can persist style-related tags and source classification information for uploaded audio. The final information available in the UI can vary. Avoid promising tempo, musical key, energy, era, or an exact style description unless the current product explicitly shows it for the selected result.When no separate style model appears
Do not assume that every catalog will produce several separate style models. The system can retain a usable primary model path even when style clustering does not yield multiple usable style results. Inspect My DNA for what was actually created.Adding more catalog material later
Use the existing primary DNA when you have more authorized catalog material. Upload the new files. Wait for the upload to become ready. Start a new processing run when no previous run is active. Review the new state in My DNA.Part 4: Use a DNA in generation
Select a DNA
Open the available DNA generation experience. Choose a DNA that is available to you. The list of available models and track references depends on that DNA and the product state.Select a model or track reference
When the UI presents both options:- Choose a model for a broader DNA starting point
- Choose a track reference for a more specific starting point
Understand context-aware controls
DNA generation controls are not identical for every model or track. They depend on the selected DNA, available model, and selected reference. Use the controls that are enabled. If a control is missing or disabled, it is not currently supported for that selection.Song and Instrumental
The current DNA generation contract accepts Song and Instrumental styles. Song refers to the available song-generation path. Instrumental refers to the available instrumental-generation path. The product may map an available reference to one of those supported request styles.Vocal references
A reference can be categorized as vocal. That does not mean the generation backend accepts a separate vocal or acapella request style. Do not promise a standalone vocal output from every vocal reference. Use the options currently enabled in the UI.Input text
When a music description or lyrics field is presented, provide text that matches the enabled flow. The exact available input fields depend on the current product selection. Keep expectations tied to the visible controls and current DNA result rather than a generic feature matrix.If generation controls change
Changing the selected DNA, model, or track reference can change the controls. This is expected context-aware behavior. Try another available reference only when it matches your intended musical starting point.Part 5: Voice DNA
Overview
Voice DNA is a dedicated training flow for one person’s voice. It has five current steps:- Voice Profile
- Upload Recordings
- Settings
- DNA Quality
- Finalize
Voice source requirements
Use solo-vocal recordings. Aim for at least 10 minutes of clean voice audio. Use one speaker or singer only. Avoid background music and heavy effects. The cleaner and more representative the source material is, the clearer the training intent.Voice consent requirement
The current Voice DNA form requires a consent checkbox. The checkbox confirms that the voice is yours or that you have the singer’s consent to clone it. The upload step cannot validate without that confirmation. Do not bypass or misrepresent the consent requirement.Step 1: Voice Profile
The Voice Profile area includes:- Creator Name
- Description
- Tags
- Cover image upload
- Avatar selection
Voice profile naming
Use a name that lets you identify the voice model later. The training flow generates a model name based on the profile context and the next voice-model number. You do not need to manually invent an internal numbered model name.Existing and primary profile context
The Voice DNA UI has a profile step. In the current primary mode, the voice model is added to the canonical primary DNA. Existing DNA content stays unchanged. Follow the profile context shown in the product when you start the flow.Step 2: Upload Recordings
Use Upload Recordings to add voice material. The form requires at least one recording. The product guidance says to aim for at least 10 minutes of clean solo-vocal audio. The two points are different:- One recording satisfies the basic form requirement
- A useful voice dataset should follow the guidance
Voice accepted formats
The current Voice DNA file input accepts:- .mp3
- .wav
- .aac
- .ogg
- .flac
- .m4a
Add existing audio files
Use the Select audio files action or drag files into the upload area. The product supports adding more files before training. Use source recordings you can keep available until upload has completed.Record in the product
The Voice DNA upload area includes a Record action. It can be used when the device and browser permit recording. Microphone permissions, device selection, and browser behavior remain device-specific. Review the resulting audio before using it as training material.Upload status
The Voice DNA uploader reports progress and status. It can report that recordings are being packaged. It can report a successful upload when recordings are ready. If it reports an upload error, read the error text and retry only after checking the files and consent.Consent before moving forward
The Voice DNA upload step validates two required conditions:- At least one recording is selected
- The consent checkbox is selected
Step 3: Voice Settings
The current Voice DNA Settings step asks who can use the voice once it is trained. It presents private and public visibility options. Use the visibility value shown in this form. Do not assume that the Full Production Personal option appears in the Voice DNA form.
Step 4: DNA Quality
DNA Quality is the Voice DNA configuration step. It uses epoch count rather than the Full Production sensitivity slider. The current range is 10 through 350 epochs. The default is 180 epochs.
Voice quality labels
The current UI labels the quality range as:- Fast at 10 epochs
- Balanced at about 180 epochs
- Studio at 350 epochs
What higher Voice quality means
The current UI describes more epochs as seeking a higher-fidelity clone while taking longer to train. Higher is not automatically better for every recording set. Use the balanced default when you do not have a source-specific reason to choose another value.Estimates in the Voice quality panel
The quality panel displays an estimated duration and token cost for the chosen epoch value. Treat those as UI estimates. They are not a guarantee of completion time, token availability, or training success.Start Voice training
Select Start Training from the DNA Quality step. The product validates the profile and upload requirements before submitting. It then moves to Finalize and starts showing the model state.Step 5: Voice Finalize
Finalize shows a summary of the Voice DNA. It displays training progress and status after the request begins. It provides a View in your profile action when the result is ready to inspect.Voice training states
The current Finalize view can show:- Queued
- Preparing
- Training
- Completed
- Failed
Queued
Queued means the Voice DNA request is waiting for available training capacity. The current UI describes this as waiting for a GPU to become available. Queued is not the same as failure.Preparing
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. This is not a reliable percentage completion estimate.
Training
When epochs are reported, the Finalize view shows training progress by current epoch and total epoch count. The percentage is tied to that reported epoch progress. Do not compare it directly to a guaranteed wall-clock completion time.Completed
Completed means the voice model reports completion. The UI changes the status to say that the voice is ready. Use View in your profile or My DNA to inspect the result.Failed
Failed means the current training result reported failure. The UI reports the failure and directs the flow back to DNA Quality so you can try again. Before retrying, review the source audio, consent, selected quality, and visible error text.Close the Voice DNA window
The current UI states that Voice DNA training continues if you close the window. The result should appear under the profile’s Voice DNAs when it is ready. Return later to inspect the product state rather than keeping the window open only to watch progress.Part 6: Visibility, moderation, and discovery
Visibility and status are separate
Full Production DNA visibility and moderation status are separate values. Visibility tells the product whether the DNA is Public, Private, or Personal. Status tells the product whether it is Pending, Approved, Rejected, or On hold. One does not replace the other.Full Production public eligibility
For the current Full Production public discovery path, a DNA must be:- Public
- Approved
Pending
Pending means the DNA has not reached the Approved state required for public discovery. Do not share it as publicly discoverable until the product shows the required public eligibility. This guide does not promise a review duration or outcome.Approved
Approved is the status required alongside Public visibility for current public discovery. It does not change the need to comply with the applicable terms and rights requirements.Rejected or On hold
Rejected and On hold are statuses distinct from Approved. Do not treat a Public visibility selection as sufficient to bypass those states. Use the status actually shown in the product.Private and Personal
Private and Personal Full Production DNA are outside public discovery. They are useful for setup, testing, and keeping the DNA out of the public browsing path. Use the product’s own visibility and status presentation when deciding how to share the DNA.Voice DNA visibility
The current Voice DNA form provides private and public options. Use its specific visibility field rather than assuming it follows the Full Production visibility set exactly.Discover
Discover shows DNA content that is eligible for public discovery. The available cards depend on the current visibility, moderation status, and product availability.
Part 7: Earn and Stripe Connect
Connect Stripe
The Earn section checks the current Stripe connection status. When no connection is present, it offers a Connect Stripe action. The action opens a Stripe Connect URL when the service returns one. If it cannot open the connection request, the UI reports an error.
Connected account view
When Stripe reports that the account is connected, the Earn page can show data from the DNA dashboard summary. The current UI can include lifetime earnings, recent earnings, usage, export data, and a path to the Stripe dashboard when that data is available. The exact dashboard values depend on the connected account and returned summary.Where creator earnings come from
The Earn history records protected creator earnings from two current sources:- Paid DNA exports: another creator completes a paid export for audio made with an eligible paid DNA.
- Fan Support: a fan completes a one-time or monthly Stripe support payment for a creator from an eligible conversation.
Paid DNA export earnings
When another creator pays to export audio made with a paid DNA, the current creator policy reserves 50% of the gross purchase payment for the DNA creator. For example, if the product records a 5.00 before any later refund or dispute adjustment. The buyer must complete the payment successfully. Opening a checkout page, closing it, or having a payment fail does not create an export earning. The creator can see the source as Paid DNA export in the Earn history once the ledger has recorded it. The price and license controls in the DNA experience can affect what is presented in the product, but they do not independently guarantee a sale, a ledger entry, a particular creator share, or a payout.Fan Support earnings
Fan Support is a direct support payment between people in an eligible Soundverse conversation. It is not a purchase of a DNA export and it does not grant ownership of the creator’s work. The current one-time Fan Support presets are 5, 25 USD. The current monthly presets are 8, and $20 USD. For each verified Fan Support payment, the current creator policy reserves 75% of the gross support payment for the recipient in the same protected creator-earnings ledger. For example, a verified 3.75 for the recipient before any later refund, dispute, or reversal adjustment. The platform records the creator’s protected earning first. It does not attempt an immediate direct transfer to the recipient’s Stripe account at the moment the fan pays.Can a creator earn before connecting Stripe?
Yes. A creator can have a protected earning recorded before they finish Stripe Connect onboarding. That lets a verified paid DNA export or Fan Support payment appear in the creator’s Earn history even when the creator has not yet linked a payout account. Until Stripe setup is complete and the account is eligible, the balance cannot be sent as a payout. The history can show a note such as Pending Stripe connection, Pending Stripe verification, or Payout setup required. Connecting Stripe later does not turn a protected balance into an instant bank transfer. The payment still needs to meet the normal protection, eligibility, threshold, and scheduling conditions described below.Connection states in Earn
Earn can show several Stripe connection states. The wording is meant to tell you what to do next:- Stripe not connected: no payout account is linked yet. Earnings can still accrue, but a payout cannot be sent until you connect Stripe.
- Setup incomplete: a Stripe account is linked, but onboarding still has steps to complete.
- Verification required: Stripe needs more information before payments or payouts can be enabled.
- Verification pending: Stripe is reviewing the submitted information.
- Payments restricted: the linked account is not currently ready for payments or payouts. Review the Stripe setup for the specific requirement.
- Stripe connected: the account is linked and the product reports it as connected. Stripe can still impose account, country, identity, banking, and payout requirements.
Read the balance cards
The cards in Earn describe different points in the creator-earnings lifecycle:- Current creator balance: the current protected creator balance recorded by the ledger after known adjustments. It is not a bank balance and is not automatically withdrawable.
- In payment protection: earnings that are still inside the payment-protection period. They are intentionally held before becoming payout-eligible.
- Eligible after protection: earnings whose protection period has ended. They still require a ready Stripe account, the minimum payout threshold, and an active payout path.
- Money accrued: the running total of creator earnings that have been recorded in the ledger.
- Money paid: the running total of creator earnings that have been recorded as paid out. It does not include money that is only protected or eligible.
- Historical paid-DNA estimate: an analytics estimate that can appear in a live dashboard. It is not a replacement for the protected ledger balance or a promise of a future payout.
Understand the earning status in history
Each Earn history row identifies a source, date, status, and amount. The current status labels mean:- Payment protection: the payment was recorded but remains inside the protection period.
- Ready for payout: the ledger marks the creator amount eligible after protection. It is still subject to payout setup and schedule checks.
- Payout processing: a payout is being processed against the ledger balance.
- Paid: the ledger has recorded the payout as completed.
- Payment review: the payment is being reviewed, such as after a refund or dispute event. The amount may be adjusted or reversed.
- Payout setup required: the earning is recorded, but Stripe connection, verification, banking, or another payout prerequisite must be completed first.
Earning-to-payout lifecycle
The normal path is:- A buyer completes a paid DNA export or a fan completes a support payment.
- The payment provider verifies the payment and the product records the creator’s share in the protected ledger.
- The earning remains in Payment protection for the applicable protection period.
- The balance becomes Ready for payout when it is eligible, provided the payout setup is complete.
- A qualifying scheduled payout moves through Payout processing and then Paid when the ledger records completion.
Manage payouts
The connected Earn view provides a way to open Stripe for payout details, banking settings, and account requirements. Use Stripe and the current product information to manage those details.Minimum payout threshold and next payout date
The live Earn view shows the current Minimum payout threshold. The current default displayed by the product is $50 USD, unless the configured account summary supplies a different amount. Reaching the threshold is necessary but not sufficient for a payout. The amount must also be eligible after payment protection, the protected ledger must be available, and the Stripe account must be ready for payouts. 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. If Earn says Not scheduled, it can mean the balance has not met the conditions yet, automatic payouts are disabled, the account still needs Stripe setup, or the product cannot currently provide a payout schedule. Stripe and the receiving bank can still take additional time to make a completed payout available. The displayed schedule is not a guarantee of bank-arrival time.When automatic payouts are disabled
Earn can continue tracking protected and eligible balances even when automatic payouts are disabled. In that state, a visible balance is not an instruction to send a transfer yourself. It means the ledger has recorded creator earnings while the payout path is not automatically sending a transfer. Check the current in-product payout message and Stripe account state rather than assuming that an eligible balance will move on a particular date.Test mode and live mode are separate
Soundverse uses separate Stripe environments for testing and live payments. In test or sandbox mode, Stripe Connect and Checkout use Stripe test objects and test funds. Test earnings can be recorded in the sandbox ledger so the Earn flow can be validated, but test payouts are disabled and no real money is sent to a bank account. In live mode, Stripe Connect and Checkout use live Stripe objects. Live earnings and live payouts are kept separate from sandbox data. Do not mix a test Stripe account, test checkout URL, or test payment method with a live flow. A checkout created in one Stripe mode is not valid in the other.Refunds, disputes, and balance changes
An Earn balance can change after a payment is first recorded. If a buyer receives a refund or a payment is disputed, the corresponding creator earning can move into Payment review or be reversed. A reversal can reduce the current balance and can affect an amount that had been waiting for payout. This protection is intentional. It prevents a creator payout from being treated as final before the source payment is settled under the applicable payment rules.Monthly Fan Support
Monthly Fan Support uses a Stripe subscription rather than a one-time payment. The current product prevents a supporter from creating duplicate active monthly support for the same creator. If an active pledge already exists, the supporter should manage that pledge instead of starting a second one. Each successful monthly payment is handled as its own verified support payment for the creator-earnings ledger. A monthly pledge can change status when it is active, past due, canceling, canceled, expired, or unpaid. Canceling a future monthly pledge does not automatically rewrite the status of support payments that were already verified. Check the current subscription and payment records in the product and Stripe.Important earnings boundary
Connecting Stripe does not itself promise:- Payout eligibility
- A royalty amount
- A payout date
- A minimum balance
- A commercial right
- A successful DNA review
What Earn does not replace
Earn is not a tax return, invoice system, tax advisor, legal advisor, or bank statement. Stripe may request identity, business, bank, or tax information for the account. Provide accurate information directly to Stripe and seek qualified professional advice for tax, accounting, licensing, or legal questions that apply to you. For a payment or payout issue, keep the non-sensitive details that help support investigate: the source shown in Earn, visible status, approximate time and timezone, amount, and any non-secret Stripe or Soundverse reference shown by the product.Part 8: Troubleshooting
Full Production upload does not start
Check that you selected at least one file. Check that each file is an accepted audio format. Check the ownership-and-authority confirmation. Read any visible error message before retrying.Full Production upload stopped partway
Keep the same source files available. Retry through the product. The current uploader can reuse a valid upload session and confirmed chunks. If the session cannot be resumed, the product can create a new session.Process DNA is unavailable
Check whether the upload has completed. Check whether the selected audio is ready to process. Check whether a previous processing run is still active. The current flow intentionally blocks duplicate starts while processing is already running.I expected a new primary DNA but do not see one
The current Full Production flow uses one primary DNA per owner. Look in My DNA for the existing primary profile and its new results. Do not expect later Full Production uploads to create a second primary profile.I do not see a new Full Production model
Check My DNA after processing has had time to complete. Check whether the current processing run is still active. Do not rely only on a model-ready email because email delivery is best effort. Do not assume every catalog must create multiple visible style models.I do not see a public DNA in Discover
Check the DNA’s visibility. Check the DNA’s moderation status. For the current Full Production public path, both Public visibility and Approved status are required.Voice DNA will not continue from upload
Confirm that you selected at least one recording. Confirm that the consent checkbox is selected. Check that the source file is an accepted audio format. Read the exact visible error before retrying.Voice DNA remains Queued
Queued indicates that the product is waiting for training capacity. It is different from a Failed result. You can close the training window and return to the profile later to check the model state.Voice DNA reports Failed
Read the Finalize status. Return to DNA Quality as the product directs. Review the source recordings, consent, quality choice, and visible error. Retry only after understanding what changed.A generation option is missing
Check the selected DNA, model, and track reference. The controls depend on that context. Use a supported selection and the controls actually enabled in the product. Do not assume that every vocal reference exposes a standalone vocal output style.Stripe Connect does not open
Read the error displayed by the Earn page. The connection flow requires a Stripe URL returned by the service. If it continues to fail, record the time, page, visible error, and whether the account was already connected. Do not share Stripe credentials or financial information in an ordinary support message.Part 9: What to send support
DNA support checklist
Include the following:- Your Soundverse account email
- Whether the issue is Full Production DNA or Voice DNA
- The exact page or step
- The visible error message
- The approximate time and timezone
- The number of selected files
- The file formats
- Whether the problem is upload, processing, visibility, generation, or Stripe Connect
Full Production issue checklist
For Full Production DNA, also include:- Whether the primary DNA already existed
- Whether the upload completed
- The selected sensitivity
- Whether a processing run was already active
- Whether the problem is in My DNA or Discover
Voice DNA issue checklist
For Voice DNA, also include:- Whether there was at least one recording
- Whether the consent checkbox was selected
- The selected epoch count
- The reported training state
- Whether the problem occurred before or after Start Training
Keep secrets private
Do not send:- Passwords
- Session cookies
- API keys
- Stripe credentials
- Raw private audio
Part 10: Keep expectations accurate
Results are catalog-dependent
Do not promise a fixed number of models, tracks, or tags before processing completes. Different source collections can produce different usable results. Inspect the primary DNA after the actual processing run.Processing time is not guaranteed
The product can show current state and progress where available. This guide does not promise a completion time for Full Production or Voice DNA training. Queueing, preparation, processing, and training are distinct states.Approval is not guaranteed
Public visibility is not a moderation approval. No statement in this guide promises a review outcome or review duration. Use the status shown in the product.Earnings are not guaranteed
Stripe connection and visible dashboard figures do not guarantee future earnings, payout timing, or royalty terms. Use the current product terms and Stripe information before making business decisions.Generation capability is context-dependent
Use the enabled DNA generation controls for the selected model or track reference. The current request contract supports Song and Instrumental. Do not promise a third universal vocal output style.Part 11: Creator planning playbooks
This section turns the product concepts above into repeatable planning paths. Use the playbooks to organize your own decisions before you upload, train, publish, generate, or connect a payout account. They do not replace the current controls in Soundverse, the applicable terms, or the rights you need for your source material.How to use these playbooks
Pick the playbook that matches the outcome you want, then work through it in order. Keep the key facts for each DNA project in a private project record: source-list summary, ownership or consent confirmation, selected settings, visible product status, and actual result. Do not put passwords, Stripe details, session cookies, API keys, bank information, or private legal documents into a normal project note. When a product state conflicts with an expectation, treat the current product state as the source of truth and use the troubleshooting sections before retrying.Playbook A: Choose the correct DNA route
Use this decision path before creating a new request.A1. Define the real creative goal
Ask what you want the system to represent. If the answer is a creator’s broader musical catalog, production identity, songs, instrumentals, or authorized acapellas, start with Full Production DNA. If the answer is one person’s voice from clean solo-vocal recordings, start with Voice DNA. Do not choose a route based only on the file extension or file name. Choose it based on the source material and the intended training goal.A2. Separate catalog questions from voice questions
Full Production DNA can include vocal material as part of a music catalog. That does not make it the voice-cloning flow. Voice DNA is dedicated to the voice-focused request and requires that the source is one authorized speaker or singer. If you need both a music catalog and a voice model, plan the source sets separately and use each flow only with the required rights and consent.A3. Check who owns or controls the source material
For Full Production DNA, make sure you have the authority to upload the catalog material. For Voice DNA, make sure the voice is yours or that you have the speaker’s or singer’s consent for this purpose. Do not treat public availability, a streaming link, a social post, or a credit line as proof of training rights. If you cannot make the product confirmation truthfully, do not submit the material.A4. Make a simple source inventory
For each planned file, record:- A clear local file name
- The original project or recording context
- The intended DNA flow
- Why the file belongs in that source set
- Whether you own it or have the needed authority or consent
- Whether it includes other voices, collaborators, samples, private speech, or sensitive material