Speech
Live text-to-speech generation, ElevenLabs instant voice cloning, and premium Qwen sidecar enrollment.
TTS 생성, ElevenLabs 음성 복제, Qwen 프리미엄 사이드카 등록을 제공하는 라이브 음성 기능입니다.
Overview
The Greit product uses catalog TTS providers through app/api/tts/generate: google, elevenlabs, openai, gemini, and edge.
Instant voice cloning is live through app/api/voices/clone and forwards voice samples to the ElevenLabs voices/add API.
Premium Qwen routes and enrollment exist under app/api/tts/qwen/* and app/api/tts/enroll/*. The Qwen sidecar is Tailscale-private today and is not yet public-tunneled.
API Reference
Generate speech audio with a selected catalog TTS provider.
| Name | Type | Required | Description |
|---|---|---|---|
| text | string | Yes | Text to synthesize. |
| engine | google | elevenlabs | openai | gemini | edge | Yes | TTS provider selected from the live Greit catalog. |
| voice | string | No | Provider-specific voice identifier. |
Response
Generated audio payload or provider-backed audio URL, depending on the selected engine.
Create an instant cloned voice through ElevenLabs.
| Name | Type | Required | Description |
|---|---|---|---|
| name | string | Yes | Display name for the cloned voice. |
| files | audio file[] | Yes | Voice sample uploads forwarded to ElevenLabs. |
Response
ElevenLabs voice metadata, including the created voice id.
Premium Qwen sidecar TTS route family.
Response
Qwen sidecar response for enrolled premium voice workflows.
Enroll premium Qwen voice workflows.
Response
Enrollment state for premium Qwen voice access.
Quickstart
curl -X POST "https://thegreit.com/api/tts/generate" \
-H "Authorization: Bearer $GREIT_SESSION_TOKEN" \
-H "Content-Type: application/json" \
-d '{"engine":"openai","text":"Preview this line","voice":"alloy"}' \
-iVersion log