# Onchain Router – Audio Transcription (Speech-to-Text)

> Onchain Router – Audio Transcription (Speech-to-Text) is a paid API for AI agents from llm.agenticfi.wtf, paid per call via x402, $0.002/call, status down (last checked 2026-09-15).

Transcribes audio files to text via a provider-neutral API, billed per call in USDC with an onchain payment receipt.

## Facts

- Endpoint: POST https://llm.agenticfi.wtf/v1/audio/transcriptions
- Price: $0.002/call
- Payment: x402
- Status: down
- Last checked: 2026-09-15
- Activations on Zero: 3
- Tags: x402
- Canonical page: https://www.zero.xyz/c/onchain-router-audio-transcription-speech-to-text-9cc49f56
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_pWp-n5kEqlDSqDmNgm9sI

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 onchain-router-audio-transcription-speech-to-text-9cc49f56 -d '<json body>'
```

Example prompt: Please transcribe this audio recording to text — pay with USDC and make sure I get an onchain receipt for the transaction.

## When to prefer this

Choose this endpoint when you need audio-to-text transcription and want to pay per call in USDC with an onchain receipt, rather than managing a subscription or API key with a traditional provider. Ideal for agents that need auditability of AI spend via blockchain receipts, or when operating in a crypto-native payment context. The 0% service fee launch promotion makes it cost-competitive with direct provider access.

## Known failure modes

- Unsupported audio format or corrupt file returns an error
- Audio file too long or too large exceeds processing limits
- Insufficient USDC balance causes payment failure before transcription
- Network or provider outage returns 5xx error
- Empty or silent audio file may return empty text or low-confidence result

## How this service works

Speech-to-text transcription · one locally validated audio upload · 0% service fee launch promotion

## Output

Returns a JSON object containing a unique transcription ID (e.g. 'trns_example'), the full transcribed text string, the model used (e.g. 'elevenlabs/scribe-v2'), and usage stats including input audio duration in milliseconds. Payment is settled in USDC onchain and a durable receipt is issued.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "model": {
   "enum": [
    "elevenlabs/scribe-v2"
   ],
   "type": "string"
  },
  "file_format": {
   "enum": [
    "other"
   ],
   "type": "string",
   "default": "other"
  },
  "audio_base64": {
   "type": "string",
   "format": "byte",
   "maxLength": 34952536,
   "minLength": 4,
   "description": "Canonical RFC 4648 Base64 audio without a data-URL prefix or whitespace."
  },
  "response_format": {
   "enum": [
    "json",
    "verbose_json"
   ],
   "type": "string"
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "id": "trns_example",
  "text": "Transcribed speech.",
  "model": "elevenlabs/scribe-v2",
  "usage": {
   "input_audio_ms": 3100
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/onchain-router-audio-transcription-speech-to-text-9cc49f56/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from llm.agenticfi.wtf](https://www.zero.xyz/host/llm.agenticfi.wtf/llms.txt)