# Venice AI Audio Transcriptions

> Venice AI Audio Transcriptions is a paid API for AI agents from api.venice.ai, paid per call via x402, $10/call, status unknown (last checked 2026-10-12).

Transcribes an uploaded audio file into text using AI speech-to-text models, optionally returning timestamps and duration.

## Facts

- Endpoint: POST https://api.venice.ai/api/v1/audio/transcriptions?utm_source=zero.xyz
- Price: $10/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-12
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/venice-ai-audio-transcriptions-dc8acef8
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_3G1xkrxhSxB9iXVn35-Z2

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-audio-transcriptions-dc8acef8 -d '<json body>'
```

Example prompt: Transcribe this audio file for me using the openai/whisper-large-v3 model — the recording is in English and I'd like word-level timestamps included in the response.

## When to prefer this

Choose this endpoint when you need server-side AI transcription via Venice AI's hosted models with x402 crypto payment support, especially when you want to select from multiple STT model backends (Whisper, Parakeet, ElevenLabs Scribe, etc.) or need word/segment-level timestamps. Prefer this over generic STT APIs when your workflow already uses Venice AI's ecosystem or you want to pay per-call with USDC without a subscription.

## Known failure modes

- Unsupported audio format returns an error — only WAV, FLAC, M4A, AAC, MP4, MP3, OGG, OGA, WEBM are accepted
- 402 Payment Required if wallet balance is insufficient for the $10 USDC per-call fee
- Language hint ignored if selected model (e.g. Parakeet) does not support language parameter
- Timestamps object absent in response if timestamps=false (default)
- Large or long audio files may time out or exceed file size limits
- Model not found error if an invalid model string is provided

## How this service works

Speech-to-text transcription.

## Output

Returns a JSON object containing the full transcribed text string, the audio duration in seconds, and optionally a timestamps object with word-, character-, and segment-level start/end times for each unit of the transcript.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "example": {
  "file": "[audio file]",
  "model": "openai/whisper-large-v3",
  "timestamps": false,
  "response_format": "json"
 },
 "properties": {
  "file": {
   "type": "string",
   "format": "binary",
   "description": "The audio file object (not a base64 string). Supported formats: WAV, WAVE, FLAC, M4A, AAC, MP4, MP3, OGG, OGA, WEBM."
  },
  "model": {
   "enum": [
    "nvidia/parakeet-tdt-0.6b-v3",
    "openai/whisper-large-v3",
    "fal-ai/wizper",
    "elevenlabs/scribe-v2",
    "stt-xai-v1"
   ],
   "type": "string",
   "default": "nvidia/parakeet-tdt-0.6b-v3",
   "example": "openai/whisper-large-v3",
   "description": "The model to use for transcription. See https://docs.venice.ai/models/overview for more information."
  },
  "language": {
   "type": "string",
   "example": "en",
   "description": "ISO 639-1 language code (e.g., \"en\", \"es\", \"fr\"). Optional - if not provided, the model will auto-detect the language. Note: Only supported by certain models (e.g., Whisper). Ignored by models that do not support language hints."
  },
  "timestamps": {
   "type": "boolean",
   "default": false,
   "example": false,
   "description": "Whether to include timestamps in the response."
  },
  "response_format": {
   "enum": [
    "json",
    "text"
   ],
   "type": "string",
   "default": "json",
   "example": "json",
   "description": "The format of the transcript output, in one of these options: json, text."
  }
 },
 "description": "Request to transcribe audio to text.",
 "additionalProperties": false
}
```

## Response schema (JSON Schema)

```json
{
 "type": "object",
 "required": [
  "text"
 ],
 "properties": {
  "text": {
   "type": "string",
   "description": "The transcribed text"
  },
  "duration": {
   "type": "number",
   "description": "Duration of the audio in seconds"
  },
  "timestamps": {
   "type": "object",
   "properties": {
    "char": {
     "type": "array",
     "items": {
      "type": "object",
      "required": [
       "char",
       "start",
       "end"
      ],
      "properties": {
       "end": {
        "type": "number"
       },
       "char": {
        "type": "string"
       },
       "start": {
        "type": "number"
       }
      }
     }
    },
    "word": {
     "type": "array",
     "items": {
      "type": "object",
      "required": [
       "word",
       "start",
       "end"
      ],
      "properties": {
       "end": {
        "type": "number"
       },
       "word": {
        "type": "string"
       },
       "start": {
        "type": "number"
       }
      }
     }
    },
    "segment": {
     "type": "array",
     "items": {
      "type": "object",
      "required": [
       "text",
       "start",
       "end"
      ],
      "properties": {
       "end": {
        "type": "number"
       },
       "text": {
        "type": "string"
       },
       "start": {
        "type": "number"
       }
      }
     }
    }
   },
   "description": "Timestamps for the transcription (only if timestamps=true)"
  }
 },
 "description": "Transcription response"
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/venice-ai-audio-transcriptions-dc8acef8/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)