# MeloTTS Multilingual Text-to-Speech

> MeloTTS Multilingual Text-to-Speech is a paid API for AI agents from wafermap.xyz, paid per call via x402, $0.002/call, status unknown (last checked 2026-10-31).

Converts text to speech audio using the lightweight MeloTTS model, supporting six languages for affordable draft narration.

## Facts

- Endpoint: POST https://wafermap.xyz/v1/tts-melo?utm_source=zero.xyz
- Price: $0.002/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-31
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/melotts-multilingual-text-to-speech-26347f88
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_axifTLpaFRxHLG_yMbg9D

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 melotts-multilingual-text-to-speech-26347f88 -d '<json body>'
```

Example prompt: Can you use MeloTTS to read this draft paragraph aloud in Spanish? The text is: 'Bienvenidos a nuestra nueva plataforma de aprendizaje en línea.'

## When to prefer this

Choose this endpoint when you need cheap, fast multilingual TTS for draft narration or prototyping across English, Spanish, French, Chinese, Japanese, or Korean — particularly when production audio quality is not yet required and cost efficiency matters. Prefer the Deepgram Aura 2 sibling endpoint on the same provider for higher-quality English-only output.

## Known failure modes

- Text exceeds 1000 character limit — request rejected with validation error
- Unsupported language code provided outside the six supported options (en, es, fr, zh, ja, ko)
- Empty or missing text field — request fails with validation error
- Service unavailable or timeout on the upstream MeloTTS model
- Payment not completed via x402 protocol — request blocked

## How this service works

MeloTTS — Light multilingual text-to-speech. Cheap narration for drafts.

## Output

Returns synthesized audio of the input text spoken in the selected language, produced by the MeloTTS lightweight model. The audio is suitable for drafts and prototyping rather than final production.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "lang": {
   "enum": [
    "en",
    "es",
    "fr",
    "zh",
    "ja",
    "ko"
   ],
   "type": "string",
   "description": "Language."
  },
  "text": {
   "type": "string",
   "maxLength": 1000,
   "minLength": 1,
   "description": "Text to speak."
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/melotts-multilingual-text-to-speech-26347f88/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from wafermap.xyz](https://www.zero.xyz/host/wafermap.xyz/llms.txt)