API Reference

API Reference#

Base URL#

The server listens on http://<host>:<port>.

POST /generate#

Generate speech from text (and optional audio) and return a WAV response.

Parameters (multipart form)#

  • text (string, required): Input prompt.

  • audio (file, optional): Input audio for STS-capable models.

  • streaming (bool, optional, default true): - true streams WAV chunks as they are produced. - false returns a single WAV file after completion.

Response#

  • Streaming mode: audio/wav with chunked transfer.

  • Non-streaming mode: audio/wav file download.

Examples#

curl -X POST "http://localhost:8000/generate" \
  -F "text=Hello world" \
  -F "streaming=true" \
  -o output.wav

curl -X POST "http://localhost:8000/generate" \
  -F "text=Hello world" \
  -F "audio=@input.wav" \
  -F "streaming=true" \
  -o output.wav

GET /health#

Health check endpoint.

Response#

JSON payload:

{"status": "healthy"}

Python API#

Package overview#

vox_serve

VoxServe: A serving system for Speech Language Models.

Key modules#

Contents