# Venice AI Voice Cloning

> Venice AI Voice Cloning is a paid API for AI agents from api.venice.ai, paid per call via x402, $10/call, status unknown (last checked 2026-10-10).

Clone a voice from an audio sample to produce a reusable voice handle for speech synthesis on Venice AI's TTS models.

## Facts

- Endpoint: POST https://api.venice.ai/api/v1/audio/voices?utm_source=zero.xyz
- Price: $10/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-10
- Activations on Zero: 0
- Tags: x402, ai_ml
- Canonical page: https://www.zero.xyz/c/venice-ai-voice-cloning-e03998d8
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_Lph3oEY5jPF__VP7pInWW

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability venice-ai-voice-cloning-e03998d8 -d '<json body>'
```

Example prompt: Clone my voice using this audio sample file so I can use it with the tts-chatterbox-hd model — I'll use it to synthesize speech that sounds like me.

## When to prefer this

Use this endpoint when you need to synthesize speech in a specific person's voice using Venice AI's TTS pipeline. It is the required first step before calling POST /v1/audio/speech with a custom cloned voice. Prefer tts-chatterbox-hd (Resemble AI) for broad format support (MP3, WAV, FLAC, M4A) and zero-shot cloning; prefer tts-minimax-speech-02-hd (MiniMax) for persistent voice templates that survive across calls (requires special access).

## Known failure modes

- HTTP 400 if the audio file format is not in the model's accepted container list (e.g. FLAC submitted to tts-minimax-speech-02-hd)
- HTTP 401/403 if the API key is invalid or the account lacks access to the selected model (tts-minimax-speech-02-hd is limited access)
- HTTP 400 if the file field is missing or malformed in multipart/form-data
- Voice handle expiry after 7 days causing 404/invalid voice errors on subsequent TTS calls
- Audio sample too short or noisy leading to poor voice quality in synthesis

## How this service works

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.

- `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 an `X-Sign-In-With-X` header for x402 wallet-based authentication. When using x402, a `402 Payment Required` response indicates insufficient balance and includes top-up instructions.

## Output

Returns a JSON object with an `id` field (a `vv_...` voice handle string) and a `model` field indicating which Venice TTS model the handle is bound to. This handle can then be passed as the `voice` parameter in POST /v1/audio/speech calls, alongside the same model, to synthesize speech in the cloned voice. Handles expire after 7 days of non-use.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "file": {
   "type": "string",
   "format": "binary",
   "description": "The voice sample audio file (multipart/form-data, field name \"file\"). Accepted containers depend on the selected model: `tts-chatterbox-hd` accepts MP3, WAV, FLAC, and M4A; `tts-minimax-speech-02-hd` accepts MP3 and WAV only. Recommended: a clean speech recording of at least 5–10 seconds."
  },
  "model": {
   "enum": [
    "tts-chatterbox-hd",
    "tts-minimax-speech-02-hd"
   ],
   "type": "string",
   "default": "tts-chatterbox-hd",
   "example": "tts-chatterbox-hd",
   "description": "The Venice TTS model the cloned voice will be paired with. The returned voice handle is only valid against this model on POST /v1/audio/speech."
  }
 },
 "description": "Request to create a cloned voice from an audio sample. The returned `vv_<id>` handle is passed back as the `voice` parameter on POST /v1/audio/speech."
}
```

## Response schema (JSON Schema)

```json
{
 "type": "object",
 "required": [
  "id",
  "model"
 ],
 "properties": {
  "id": {
   "type": "string",
   "example": "vv_voice_abc123xyz",
   "description": "The cloned voice handle. Pass this to POST /v1/audio/speech as the `voice` parameter alongside the same `model` used to create it. Handles expire automatically after the per-model retention window (currently 7 days for both supported models); for `tts-minimax-speech-02-hd` each successful TTS request resets that window."
  },
  "model": {
   "type": "string",
   "example": "tts-chatterbox-hd",
   "description": "The Venice TTS model this voice can be used with."
  }
 },
 "additionalProperties": false
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/venice-ai-voice-cloning-e03998d8/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from api.venice.ai](https://www.zero.xyz/host/api.venice.ai/llms.txt)