# ForgeMesh Voice Long-Form Text-to-Speech

> ForgeMesh Voice Long-Form Text-to-Speech is a paid API for AI agents from voice.forgemesh.io, paid per call via x402, $0.003/call, status unknown (last checked 2026-09-14).

Converts long-form text into spoken audio using persona voices across 31 languages, with pay-per-call x402 micropayment billing and an OpenAI-compatible API shape.

## Facts

- Endpoint: POST https://voice.forgemesh.io/v1/audio/speech-long
- Price: $0.003/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-14
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/forgemesh-voice-long-form-text-to-speech-f40f9edc
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_1UtoOucJqANy9e2SZqiw4

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-voice-long-form-text-to-speech-f40f9edc -d '<json body>'
```

Example prompt: Read the following blog post out loud in English using a warm narrator voice and give me the audio as a WAV file: 'Welcome to ForgeMesh. Today we explore how AI agents can speak to the world in 31 languages with just a single API call.'

## When to prefer this

Choose this endpoint when you need long-form text-to-speech synthesis with persona voices, multilingual support (31 languages), and an OpenAI-compatible request shape — especially in agentic workflows where pay-per-call x402 micropayments are preferred over subscription billing.

## Known failure modes

- Payment not included or insufficient USDC — returns 402 Payment Required
- Unsupported response_format value — returns 400 Bad Request
- Text input missing or empty — returns 400 Bad Request
- Requested voice name not found — may return 400 or fall back to default
- Service temporarily unavailable — returns 503

## How this service works

OpenAI-compatible text-to-speech endpoint (input, voice, response_format) for drop-in use with existing OpenAI TTS clients, 1 of 10 standard voices, for 501-2000 characters, WAV/FLAC/OGG output. Use it to: voice a longer OpenAI-formatted request without an API key, migrate an existing TTS integration to pay-per-call pricing, synthesize extended text through a familiar request shape. USDC on Base via x402.

## Output

An audio file (WAV, FLAC, or OGG) containing the synthesized speech of the provided text, returned as an OpenAI-compatible audio response with the appropriate content-type header.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "input": {
   "type": "string",
   "maxLength": 2000,
   "minLength": 1,
   "description": "OpenAI-compatible text input to synthesize, max 2000 characters"
  },
  "model": {
   "type": "string",
   "description": "Ignored — always uses ForgeMesh Voice"
  },
  "voice": {
   "enum": [
    "M1",
    "M2",
    "M3",
    "M4",
    "M5",
    "F1",
    "F2",
    "F3",
    "F4",
    "F5"
   ],
   "type": "string",
   "description": "Standard voice: M1-M5 or F1-F5"
  },
  "response_format": {
   "enum": [
    "wav",
    "flac",
    "ogg"
   ],
   "type": "string",
   "description": "Requested audio format"
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "description": "OpenAI-compatible inline speech audio response",
  "content_type": "audio/wav"
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/forgemesh-voice-long-form-text-to-speech-f40f9edc/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from voice.forgemesh.io](https://www.zero.xyz/host/voice.forgemesh.io/llms.txt)