/api/v1/audio/voices

Clone a voice from an audio sample. Returns a voice handle (vv_...) that can be passed to POST /v1/audio/speech as the voice parameter to synthesize speech in the cloned voice; the handle must be paired with the same model used to create it.

Supported models:

• tts-chatterbox-hd (Resemble AI) — accepts MP3, WAV, FLAC, and M4A samples. Zero-shot: no persistent voice template is derived; the reference audio is uploaded to a TTL’d storage object and re-read on each synthesis call. Handles expire automatically after 7 days. Available to regular users.

• tts-minimax-speech-02-hd (MiniMax) — accepts MP3 and WAV only. Persistent: the upstream provider derives a custom voice template that survives across synthesis calls. Auto-deleted after 7 days without use; each successful TTS request resets the window. Currently limited access — contact support@venice.ai if you need it enabled.

Samples in containers outside the per-model allowlist are rejected with HTTP 400 before any upload.

Authentication: This endpoint accepts either a Bearer API key or a SIGN-IN-WITH-X header for x402 wallet-based authentication. The legacy X-Sign-In-With-X header is also accepted during migration. When using x402, a 402 Payment Required response indicates insufficient balance and includes top-up instructions.