Labels protect their artists. TrustMark adds the voice-level authorization layer.
Labels and studios already protect their artists through contracts, representation, and legal teams. Those relationships work. But AI voice cloning operates outside those relationships — generating convincing audio from a 30-second public sample, dubbing voices into unauthorized languages, or training on publicly available recordings. TrustMark gives labels and studios the technical complement to their existing protections: Voice Module authorization, invisible audio watermarks, and a permanent provenance trail that proves what was authorized and by whom.
30s
Sample needed to clone a voice
40+
Languages AI can dub a voice into
$26B
Global recorded music revenue (2024)
What TrustMark Does For Music & Voice
Voice Module Consent
A voice is registered as a separate identity module with independent authorization controls. Authorizing face rendering does not authorize voice cloning. Every module requires explicit, separate authorization.
TRAINING_USE Rights Flag
A dedicated rights flag that must be explicitly granted before a voice can be used to train any AI model. Without it, voice data cannot legally enter any training pipeline under TrustMark authorization.
Sound IDs & Audio Watermarks
Every authorized audio output gets a unique Sound ID and an AudioSeal watermark embedded in an inaudible frequency range — detectable by TrustMark’s verification system. Audio verification works without the original file.
ScriptModule — For Songwriters & Lyricists
Lyrics, compositions, and written works get the same protection as voice. ScriptModule registers written works with a unique fingerprint. Training rights are always a separate right — a label or platform requesting lyrics for analysis does not get the right to train a model on them. Every access is logged permanently.
Emergency Revocation — Shared Safety Mechanism
Emergency revocation stops unauthorized voice cloning and AI audio generation — it never affects valid licenses. If a label has a two-year dubbing authorization, that deal is honored. It targets the threats outside valid agreements: the AI company cloning a voice without a contract, the foreign platform generating songs without authorization. Anyone in the authorized chain — artists, agents, labels, guild reps — can trigger it when they detect unauthorized use.
Build On Our Infrastructure
Labels, distributors, and music platforms can integrate TrustMark directly into their own ecosystems using the open API.
How It Works For Music & Voice
Register a voice module
Self-service via phone mic or studio-quality recording. The voice is processed into a secure identity record. The raw recording is never stored.
Set per-project authorization
Each dubbing project, AI song generation, or voice synthesis request requires a separate token. You specify which rights are granted: VOICE_RENDER, VOICE_DUBBING, VOICE_CLONE — and explicitly whether TRAINING_USE is permitted.
Monitor and verify
Every authorized audio output carries a Sound ID. TrustMark2019s verification system can detect watermarks in any audio file 2014 even without the original.
Label secures English dubbing rights. Later expands to Spanish.
Voice Module registered, VOICE_DUBBING authorized for English territory
English dubbing token issued, receipts generated and logged
Label requests Spanish territory expansion with updated compensation
Agent reviews terms and approves on behalf of the artist
Both territories fully documented — clean audit trail for label and agent
What Labels & Studios Get
- Voice Module with independent authorization controls
- Training rights always separate — explicit opt-in required
- Sound IDs on every authorized audio output
- Invisible watermarks — inaudible, codec-resistant
- ScriptModule for songwriters — lyrics and compositions protected
- Integrates into existing label and distributor workflows