Skip to content

EXAMPLE.md

Latest commit

 

History

History
49 lines (33 loc) · 925 Bytes

EXAMPLE.md

File metadata and controls

49 lines (33 loc) · 925 Bytes
Raw

Docker Compose Server

Run the audio.cpp TTS server (as configured in server.json) in Docker.

Setup

1. Download the PocketTTS model

Get the English model from kyutai/pocket-tts on Hugging Face. Place the files from languages/english/ into ../models/pocket-tts/languages/english/:

The directory should look like:

../models/pocket-tts/languages/english/
├── model.safetensors
├── tokenizer.model
└── embeddings/
    ├── alba.safetensors
    └── ...

2. Start the server

CPU:

docker compose -f cpu-server.yml up

GPU (CUDA):

docker compose -f cuda-server.yml up

3. Optionally: Wait for server to be ready

./wait-for-server.sh

Generate speech

./tts.sh

This sends a request to http://localhost:8080/v1/audio/speech and saves the result to output/speech.wav.