# SecretVoice API

> Expressive, low-latency text-to-speech tuned for closeness — 29 voices, 21 languages, synchronous / streaming / realtime WebSocket delivery.

Programmatic access to the SecretVoice text-to-speech API. Drop-in compatible with the ElevenLabs protocol.

## Contact
- Website: https://secretvoice.dev
- API docs: https://secretvoice.dev/docs
- Support: support@secretvoice.dev

## Key resources
- [API Reference (docs)](https://secretvoice.dev/docs)
- [OpenAPI 3 Specification](/openapi.json)
- [Full reference for agents (llms-full.txt)](/llms-full.txt)

## Core endpoints
- `POST /v1/text-to-speech/{voice_id}` — convert text to speech (audio file)
- `POST /v1/text-to-speech/{voice_id}/stream` — streaming audio as it's synthesized
- `WS wss://ws.secretvoice.dev/v1/text-to-speech/{voice_id}/stream-input` — realtime speech (voice agents)
- `GET /v1/voices` — list the 29 voices
- `GET /v1/models` — list models (`secret-v1`)
- `GET /v1/history` — your generation history
- `GET /v1/user` / `GET /v1/user/subscription` — account & balance

## Authentication
Send your API key in the `xi-api-key` header:
```
xi-api-key: $SECRETVOICE_API_KEY
```

## Quickstart
```bash
curl -X POST https://api.secretvoice.dev/v1/text-to-speech/velvet \
  -H "xi-api-key: $SECRETVOICE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"text": "Come closer. Let me tell you a secret.", "model_id": "secret-v1"}' \
  --output voice.mp3
```

## Billing
Pay per audio-second of produced content. Balance is tracked live in `/v1/user/subscription`.