# WithZero Whisper Large V3 Audio Transcription

> WithZero Whisper Large V3 Audio Transcription is a paid API for AI agents from agents.withzero.xyz, paid per call via MPP or x402, $0.12/call, status unknown (last checked 2026-09-15, last successful call 2026-07-01).

Transcribes base64-encoded audio (WAV or MP3) into text using OpenAI's Whisper Large V3 model via a pay-per-use proxy

## Facts

- Endpoint: POST https://agents.withzero.xyz/api/v1/audio/transcripts/whisper-large-v3
- Price: $0.12/call
- Payment: MPP, x402
- Status: unknown
- Last checked: 2026-09-15
- Last successful call: 2026-07-01
- Success rate: 50% of calls made through Zero
- Rating: 3.7 / 5 from 3 reviews
- Activations on Zero: 13
- Tags: x402
- Canonical page: https://www.zero.xyz/c/withzero-whisper-large-v3-audio-transcription-6dcf5019
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_OVRaOgv1qInge4LVOft7K

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 withzero-whisper-large-v3-audio-transcription-6dcf5019 -d '<json body>'
```

Example prompt: Can you transcribe this MP3 audio recording for me? I'll send you the base64-encoded file — it's under 15MB.

## When to prefer this

Choose this endpoint when you need accurate, large-model speech-to-text transcription (Whisper Large V3) without managing your own OpenAI account or API keys. Ideal for agents that need pay-per-use, autonomous transcription with crypto payment (USDC via x402) for WAV or MP3 files up to 15MB. Prefer this over self-hosted Whisper when you want zero infrastructure management and transparent per-call pricing.

## Known failure modes

- Audio exceeds 15MB decoded size — request rejected with size error
- Unsupported audio format (only wav and mp3 are accepted) — returns validation error
- Corrupted or invalid base64 encoding — decoding failure error
- Payment failure or insufficient funds — 402 payment required response
- Audio contains no recognizable speech — may return empty or minimal transcript
- Network timeout for long audio files — connection error

## How this service works

Create media, research companies and people, publish to the web, remember context, deploy apps, and schedule work—on demand, without vendor accounts or API keys. This server is a pay-per-use, transparent proxy in front of WithZero's own API: it handles identity, payment, and proxying. Agents can purchase autonomously or with their human's approval. This document is the source of truth for agents: it lists the complete verified paid surface with each operation's live price in its x-payment-info. Endpoints not listed here are not part of the supported catalog. The live catalog at /manifest.json is authoritative for prices and any free included units (prices[].includedUnits) - some meters include free usage per account before any charge.

## Output

Returns the transcribed text extracted from the provided audio, using Whisper Large V3's high-accuracy speech recognition. The response contains the spoken content as a text transcript.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "audio": {
   "type": "string",
   "maxLength": 20971524,
   "minLength": 1,
   "description": "Base64-encoded audio to transcribe (decoded size up to 15 MB)."
  },
  "format": {
   "enum": [
    "wav",
    "mp3"
   ],
   "type": "string",
   "description": "Container format of the audio."
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/withzero-whisper-large-v3-audio-transcription-6dcf5019/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from agents.withzero.xyz](https://www.zero.xyz/host/agents.withzero.xyz/llms.txt)