# OpenRouter Speech Synthesis

> OpenRouter Speech Synthesis is a paid API for AI agents from OpenRouter at openrouter.mpp.tempo.xyz, paid per call via MPP, price unknown, status unknown (last checked 2026-10-29).

Converts text to speech audio bytes using configurable models and voices, billed per character or audio unit

## Facts

- Endpoint: POST https://openrouter.mpp.tempo.xyz/v1/audio/speech?utm_source=zero.xyz
- Price: price unknown
- Payment: MPP
- Status: unknown
- Last checked: 2026-10-29
- Activations on Zero: 0
- Provider: OpenRouter
- Docs: https://openrouter.ai
- Website: https://openrouter.ai
- Tags: llm, unified, multi-model, chat, ai
- Canonical page: https://www.zero.xyz/c/openrouter-speech-synthesis-58fd0286
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_S_fSy6twAV6eT41WOmmam

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 openrouter-speech-synthesis-58fd0286 -d '<json body>'
```

Example prompt: Using OpenRouter's speech endpoint, read this paragraph aloud in a calm, authoritative female voice: 'Welcome to our annual report. This year marked record growth across all divisions.' — use the best available neural TTS model and return the audio bytes.

## When to prefer this

Choose this endpoint when you need high-quality text-to-speech audio generation through a unified multi-model API with live pricing, and want the flexibility to select from multiple TTS models and voices. Prefer this over single-provider TTS APIs when you want model choice, transparent per-character or per-audio pricing, and access to OpenRouter's model catalog without managing multiple provider credentials.

## Known failure modes

- Invalid or unsupported voice identifier returns an error
- Unsupported model ID causes a 400 or 422 error
- Text input exceeding model character limits causes rejection
- Missing required fields (text, model, voice) returns a validation error
- Insufficient USDC balance causes payment failure
- Network timeout for long text inputs
- Model temporarily unavailable returns a 503 error

## How this service works

Generate speech/audio bytes (one-shot charge; character or bounded audio pricing; models, prices and voices at GET /v1/audio/models)

## Output

Raw audio bytes representing the synthesized speech, ready for playback or storage. The audio format and encoding depend on the selected model and request parameters. Pricing is determined per character or per unit of audio as specified by the selected model's pricing tier.

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/openrouter-speech-synthesis-58fd0286/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from openrouter.mpp.tempo.xyz](https://www.zero.xyz/host/openrouter.mpp.tempo.xyz/llms.txt)