AI Media
Voice clone / TTS
ElevenLabs-class VO. Cloned voices require consent and license — that gate is part of the plan, not an afterthought.
≈4s
Listen
Output stage
Generated audio lands here when a provider completes. Until then this stage stays empty on purpose — no fake waveforms, no implied live render.
clonedAuren (studio)
Auren (studio)
Primary TSWS narration — consent + license required
Consent gate
Cloned voice work requires an explicit consent record and license before run. The job runner will block until that condition is met. That is the product.
Honesty mark: mock provider paths never invent audio. Simulated runs stay labeled. Live requires credentials.
Brief
Cloned voice: attach the consent record and licence before running. The gate is the point.