# Microsoft AI Speech (Text-to-Speech via OpenRouter)

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

Converts input text to synthesized audio bytes using Microsoft AI speech models, billed per character or audio unit via one-shot USDC payment.

## Facts

- Endpoint: POST https://microsoft-ai.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: Microsoft AI
- Docs: https://openrouter.ai
- Website: https://openrouter.ai
- Tags: ai
- Canonical page: https://www.zero.xyz/c/microsoft-ai-speech-text-to-speech-via-openrouter-9b4876d1
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_3zID1P36AMKrUcemL2Xoj

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 microsoft-ai-speech-text-to-speech-via-openrouter-9b4876d1 -d '<json body>'
```

Example prompt: Can you generate a spoken audio clip of this text using a natural-sounding Microsoft AI voice: 'Welcome to our platform. We're glad to have you here.' — use a professional female voice if available.

## When to prefer this

Choose this endpoint when you need to synthesize speech using Microsoft AI voice models accessible via OpenRouter, with per-character or bounded-audio pricing paid via one-shot USDC. Ideal for agents that need programmatic, pay-as-you-go text-to-speech without a subscription, especially when Microsoft voice quality or Azure-backed models are preferred over alternatives like ElevenLabs or Google TTS.

## Known failure modes

- Invalid or unsupported model ID returns a 400 or 404 error
- Text input too long for the selected model's character limit
- Insufficient USDC balance for one-shot payment results in payment failure
- Unsupported voice name for the selected model
- Malformed request body returns validation error
- Network timeout for very long text synthesis jobs

## 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

Returns raw synthesized audio bytes representing the spoken version of the input text. The audio format (e.g. MP3, WAV, Opus) depends on the model and request parameters. The response can be saved directly as an audio file or streamed to a client.

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/microsoft-ai-speech-text-to-speech-via-openrouter-9b4876d1/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from microsoft-ai.mpp.tempo.xyz](https://www.zero.xyz/host/microsoft-ai.mpp.tempo.xyz/llms.txt)