# ForgeMesh Voicemail & Audio Transcription

> ForgeMesh Voicemail & Audio Transcription is a paid API for AI agents from x402.forgemesh.io, paid per call via x402, $0.03/call, status unknown (last checked 2026-09-15).

Transcribes spoken audio files (up to ~20 minutes, max 25MB) into written text with automatic language detection across ~99 languages, discarding files immediately after processing.

## Facts

- Endpoint: POST https://x402.forgemesh.io/voicemail-transcription
- Price: $0.03/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-15
- Activations on Zero: 1
- Tags: x402
- Canonical page: https://www.zero.xyz/c/forgemesh-voicemail-audio-transcription-54657dc9
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_YAGT1gSdWre4TYU6bMFmT

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 forgemesh-voicemail-audio-transcription-54657dc9 -d '<json body>'
```

Example prompt: Can you transcribe this voicemail for me? The audio file is at https://storage.example.com/voicemail-2024-06-10.mp3 — just convert it to text and auto-detect whatever language it's in.

## When to prefer this

Use this endpoint when you need to convert a spoken audio file (voicemail, podcast clip, recorded call) to text via a simple URL submission, especially when privacy matters since files are discarded post-transcription. It supports ~99 languages with auto-detection, making it ideal for multilingual pipelines. Prefer this over general-purpose STT services when working within an x402 micropayment-based agent workflow at $0.03 per transcription.

## Known failure modes

- Audio file URL is not publicly accessible or returns a 403/404 — transcription fails
- Audio file exceeds 25MB size limit — rejected before processing
- Audio duration exceeds ~20 minutes — may be rejected or truncated
- Unsupported audio format — transcription fails
- Network timeout fetching the remote audio URL
- Audio quality too poor for accurate transcription — low-confidence or garbled output

## How this service works

Voicemail and call transcription: converts spoken audio up to about 20 minutes long into written text, auto-detecting the language among nearly 99 supported. Files are discarded right after transcription completes, nothing is retained. Useful for logging voicemails, transcribing podcast segments, or feeding spoken content into downstream text-processing agents.

## Output

A written text transcript of the spoken audio, with automatic language detection applied. The source audio file is discarded immediately after transcription completes, so no audio is retained on the server.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "audio_url": {
   "type": "string",
   "description": "Public URL of an audio file, max 25MB"
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "text": "The quick brown fox jumps over the lazy dog. ForgeMesh utility grid speech fixture."
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/forgemesh-voicemail-audio-transcription-54657dc9/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from x402.forgemesh.io](https://www.zero.xyz/host/x402.forgemesh.io/llms.txt)