HISTORICAL DIG · JUL—AUG 2026
The sample plays. Which rights still need checking?
Platforms now separate vocal impersonation, content mismatch, spam, and AI contribution disclosures. Current Chinese judicial guidance asks whether a voice identifies a particular person and how it is used, while the U.S. Copyright Office treats digital replicas and copyrightability as distinct questions. “AI-generated,” “in a style,” and “the tool exported it” are not a single rights status.
HOW IT RELATES TO WORKFLOW 001
Relation to the current subject
It follows the first listen without turning “understood” into “ready to release.”
The previous historical record asks one person who has not seen the lyrics to hear one change and decide whether to continue, change one thing, or stop. This record handles a different gate: before further mixing or distributor delivery, how can unknowns in source material, identifiable voice, tools and contracts, intended use, and credits remain visible? It shares Workflow 001's method—freeze an inspectable result before adding tools and cost—but it is not the same topic.
Record state at the time
- AI audio and music
- Voice and source-material rights check
- Standard record
- Developing
- Historical E1 · platform, judicial, and maker sources rechecked · no BitShovel audio or rights review
THE GUIDANCE AT THE TIME
Judgment at the time
“Recorded” does not mean “permitted to release.” Unknown means hold.
Open full rationale and return rules
Freeze one candidate master and stop further production. Give lyrics, melody, voice, samples, tool and model, and human edits separate rows. Record provenance, relevant people or rightsholders, intended use, platform and territory, term, credit or AI-contribution fields, and an openable receipt. Use only recorded, replace-or-hold, and unknown. Recorded still does not mean permission to release. If a voice may identify a particular person and you cannot show permission scoped to the intended use, hold public release. That is a conservative operating gate—not an infringement ruling.
- Who it was for
- People with a listenable sample who are considering further mixing, a collaborator handoff, or distribution but have not recorded every input, voice identity, intended use, and receipt
- Why it mattered then
Open note
The later provenance is checked, the more likely a voice, sample, or permission scope must be replaced after mixing, mastering, artwork, delivery, and promotion already consume effort. The expensive step is not making a sheet; it is adding production while permission remains unknown.
- Where the judgment stops
Open note
BitShovel has no song, voice file, contract, identity test, distributor account, distributor confirmation, or legal review here. Spotify material describes its platform policy and current disclosure direction. The Supreme People's Court guidance and U.S. Copyright Office report are jurisdiction-specific interpretive and policy materials that cannot be merged into a worldwide conclusion. The pinned ACE-Step version shows audio upload and editing surfaces; it grants no permission for uploaded material. The master, digest, voice preset, states, and decision in the explainer are fixed fictional data. The hold sheet is not rights clearance, contract review, legal advice, or platform preapproval.
CHECK BEFORE READING ON
Fit check
First confirm that a blocker list is an acceptable result and every unknown can have an owner.
This record may help
- You have one candidate sample and can locate its project, drafts, input files, and the tool or model version used.
- You can accept a blocker list as the whole result and will not publish or deliver while any required item remains unknown.
Do not use it yet when
- A complaint, takedown, contract dispute, or impersonation allegation already exists and requires an actual legal or platform process.
- Release is imminent, the voice may identify a particular person, and no written permission or reachable person or rightsholder exists.
- You want a self-completed sheet to prove originality, commercial clearance, worldwide release, or platform acceptance automatically.
UNRUN, UNFROZEN 45-MINUTE PRE-RELEASE ORGANIZATION
Historical trial
Freeze one candidate master, route unknowns, and never stamp your own permission to release.
- Time needed
- A 45-minute pre-release organization cap. If anything remains unknown, hold; legal and distributor waiting time is outside the cap
- Likely cost
- Add no generation, mixing, distribution, or promotion spend; organize existing material and public rules only
- Permission boundary
- Review existing drafts, projects, and contract copies read-only; upload no voice, sample, lyric, client material, account, or credential
- 01
First step
Open this step
Copy the candidate master into a read-only file and record its date and digest. Give lyrics, melody, voice, samples, tool and model, and human edits separate rows. Add provenance, relevant people or rightsholders, permission or contract basis, intended use, platform and territory, term, credit and AI-contribution fields, and receipt location. Mark only recorded, replace-or-hold, or unknown. Route every unknown to a person or qualified channel that can answer it; without an owner, the hold remains.
- 02
What should exist
Open this step
A traceable release hold sheet showing what has a record, what must be replaced, and what needs a distributor or qualified reviewer—not proof that the work may be released.
- 03
How to tell it worked
Open this step
Every input in the master traces to a source or an explicit unknown, and identifiable voice, use, territory, term, platform fields, and credits each have an owner. Success means blockers are visible and routable—not that rights are cleared or release is approved.
- 04
Stop when
Open this step
Stop when voice identity, lyrics, melody, samples, tool terms, or use scope is unclear; a real dispute already exists; or continuing would expose private contracts, accounts, or client material.
- 05
How to back out
Open this step
Freeze the current sample and sheet; do not publish or deliver. Replace one disputed element while retaining before-and-after versions, reason, and new receipt. Route consequential or disputed items to the target platform, distributor, or a qualified professional.
BITSHOVEL R3 RELEASE HOLD · FIXED FICTIONAL RECORD
Method and interfaces
Provenance, identifiable voice, intended use, and delivery fields stay separate; one unknown holds release.
Open method boundary
The master, digest, lyric and melody provenance, voice preset, terms state, and release decision are demonstration data with no matching song, person, contract, or platform account. Green means only “record present,” never rights cleared.
The master, digest, lyric and melody provenance, voice preset, terms state, and release decision are fixed fictional examples. Green means only “record present,” never permission to release.
About this image and its source
BitShovel explainer created August 16, 2026; it contains no real song, voice, lyrics, person, contract, account, tool run, rights judgment, or release result.
Open source page ↗SOURCES AND EVIDENCE
Sources and evidence
As of August 16, 2026, the external material separately supports one platform rule, voice and digital-replica boundaries in two jurisdictions, and one pinned tool surface. It cannot be merged into worldwide permission to release.
Open evidence boundary
Revision 3 keeps the historical E1 label but gives each external source one job. Spotify supports one platform rule; the Chinese judicial guidance supports identity and context as relevant factors; the U.S. Copyright Office supports jurisdictional gaps around digital replicas; and ACE-Step supports only pinned upload and editing surfaces. The BitShovel explainer makes the reason for a hold visible. None replaces a real contract, permission, release field, or qualified review.
- 01Platform policy · Rechecked
Spotify separates unauthorized vocal impersonation from AI contribution disclosure
Open source boundary
Spotify's current announcement says vocal impersonation is allowed only when the impersonated artist authorizes it, while AI vocals, instrumentation, and post-production are disclosed through labels, distributors, or partners. This is Spotify's platform handling—not another service's rule, a contract, or a legal conclusion.
Open primary source or explainer ↗ - 02Chinese judicial guidance · Rechecked
Supreme People's Court guidance emphasizes voice identifiability and context of use
Open source boundary
The guidance connects natural-person voice protection to Civil Code personality-right rules and asks whether the voice identifies a particular person, while cautioning against treating simple imitation mechanically. It is a study and research reference, not a ruling on a work, permission, or dispute.
Open primary source or explainer ↗ - 03U.S. policy report · Rechecked
The U.S. Copyright Office treats digital replicas as a separate issue and records gaps across existing protections
Open source boundary
The report addresses digital replicas that realistically reproduce a person's voice or appearance and recommends new federal protection. It shows why one U.S. copyright rule, platform policy, or self-completed sheet cannot produce a worldwide conclusion.
Open primary source or explainer ↗ - 04Pinned maker material · Rechecked
Pinned ACE-Step exposes audio upload, voice cloning, lyric editing, and remix surfaces
Open source boundary
The maker also asks users to verify originality, disclose AI involvement, and obtain appropriate permissions. Neither features nor disclaimers automatically license user inputs, an impersonated person, or outputs. The old node image includes a demo-audio filename with unestablished provenance, so it is not redistributed as a BitShovel content visual.
Open primary source or explainer ↗ - 05BitShovel Revision 3 explainer · Rechecked
The release hold sheet places provenance, intended use, and unresolved items in one fixed fictional record
Open source boundary
No audio, person, or contract corresponds to the image. It establishes only that BitShovel expressed “why not release” as a reviewable page structure—not that rights, platform acceptance, or the method is valid.
Open primary source or explainer ↗
REVISION RECORD
Revision record
Revision 3 stops turning “sounds like an artist” into an automatic result—and stops a sheet from impersonating rights clearance.
Open 3 revisions
- Revision 1
Separated voice identity, lyrics, melody, samples, model use, and human edits into a pre-release rights sheet.
- Revision 2
Replaced the text-heavy node graph with an openly licensed studio photograph and clarified that it was not a BitShovel recording or permission receipt.
- Revision 3
Removed the atmospheric studio photo and node graph containing a demo-audio filename with unestablished provenance, added a fixed fictional release hold sheet, and separated platform rules, voice identifiability, jurisdiction, contracts, and delivery fields.
- What the record said before
- Record provenance, permission, and credit; stop when a voice sounds like a named artist without permission, then use one sheet to mark continue, replace, or unknown.
- What changed
- BitShovel cannot turn “sounds like” into an automatic cross-platform, cross-jurisdiction infringement result. The record now asks whether the voice may identify a particular person and whether permission matches the use. Any unknown holds public release and routes the issue to its owner. The sheet routes questions; it never returns “rights cleared” or “may release.”
WHAT THIS RECORD LEAVES
What this record leaves
The next move is not more mixing. It is giving every unknown someone who can answer it.
A release hold sheet moves rework forward: replace the disputed voice, obtain scoped permission, recheck current distributor fields, or route an actual dispute to the target platform, distributor, and a qualified professional. Without an owner, receipt, or explicit scope, the hold remains.
Back to the Dig index