What is chatterbox-tts-api?

Chatterbox TTS API is a FastAPI-powered REST server that provides an OpenAI-compatible text-to-speech endpoint with voice cloning capabilities. Built on top of the Chatterbox TTS model from Resemble AI, it enables users to generate speech in 22 languages using their own voice samples. The API includes a React-based web UI for managing voice libraries and testing generation, real-time audio streaming (both raw and SSE formats), long-text synthesis with background job processing, and comprehensive status monitoring. It can be deployed via Docker Compose (with GPU, CPU, and uv-optimized variants) or run locally with Python using uv or pip. The project is licensed under AGPL-3.0 and integrates with Open WebUI as a drop-in TTS backend.

Should you self-host chatterbox-tts-api?

chatterbox-tts-api scores 53/100 on selfhost.directory (how scoring works) and ranks #6 of 9 in Text to Speech. It has 694 GitHub stars.

There was no stable release in the past 12 months, so check the repository's recent activity before you rely on it.

Setup is moderate: expect some configuration beyond a single command. It is AGPL-3.0-licensed and written mostly in Python.

Its closest alternative on selfhost.directory is OmniVoice-Studio (94/100, 91 965 stars). It scores higher overall, so compare the two before you pick. See chatterbox-tts-api vs OmniVoice-Studio side by side.

Features

Auto API documentation
Long text synthesis
Memory Management
Multilingual support (22 languages)
OpenAI Compatible API
Parameter control
React web UI
Real-time Audio Streaming
Smart text processing
Status Monitoring
Type-safe API
Voice Cloning
Voice library management

How to install chatterbox-tts-api

Install chatterbox-tts-api with Docker Compose

Environment variables

PORT, FRONTEND_PORT, HOST, CORS_ORIGINS, DEVICE, VOICE_SAMPLE_PATH, MODEL_CACHE_DIR, VOICE_LIBRARY_DIR, USE_MULTILINGUAL_MODEL, EXAGGERATION, CFG_WEIGHT, TEMPERATURE, MAX_CHUNK_LENGTH, MAX_TOTAL_LENGTH, LONG_TEXT_DATA_DIR, LONG_TEXT_MAX_LENGTH, LONG_TEXT_CHUNK_SIZE, LONG_TEXT_SILENCE_PADDING_MS, LONG_TEXT_JOB_RETENTION_DAYS, LONG_TEXT_MAX_CONCURRENT_JOBS, MEMORY_CLEANUP_INTERVAL, CUDA_CACHE_CLEAR_INTERVAL, ENABLE_MEMORY_MONITORING

Clone and start with Docker Compose

git clone https://github.com/travisvn/chatterbox-tts-api cd chatterbox-tts-api

Use Docker-optimized environment variables

cp .env.example.docker .env # Docker-specific paths, ready to use

Or: cp .env.example .env # Local development paths, needs customization

Choose your deployment method:

API Only (default)

docker compose -f docker/docker-compose.yml up -d # Standard (pip-based) docker compose -f docker/docker-compose.uv.yml up -d # uv-optimized (faster builds) docker compose -f docker/docker-compose.gpu.yml up -d # Standard + GPU docker compose -f docker/docker-compose.uv.gpu.yml up -d # uv + GPU (recommended for GPU users) docker compose -f docker/docker-compose.cpu.yml up -d # CPU-only docker compose -f docker/docker-compose.blackwell.yml up -d # Blackwell (50XX) NVIDIA GPUs

API + Frontend (add --profile frontend to any of the above)

docker compose -f docker/docker-compose.yml --profile frontend up -d # Standard + Frontend docker compose -f docker/docker-compose.gpu.yml --profile frontend up -d # GPU + Frontend docker compose -f docker/docker-compose.uv.gpu.yml --profile frontend up -d # uv + GPU + Frontend docker compose -f docker/docker-compose.blackwell.yml --profile frontend up -d # (Blackwell) uv + GPU + Frontend

Watch the logs as it initializes (the first use of TTS takes the longest)

docker logs chatterbox-tts-api -f

Test the API

curl -X POST http://localhost:4123/v1/audio/speech
-H "Content-Type: application/json"
-d '{"input": "Hello from Chatterbox TTS!"}'
--output test.wav

Multiple variants: standard, uv-optimized, GPU, CPU, Blackwell; optional frontend profile

How to configure chatterbox-tts-api

CFG_WEIGHT0.5CFG weight for pace control
CFG weight for pace control
values: 0.0-1.0
CORS_ORIGINS*CORS origins (comma-separated list)
CORS origins (comma-separated list)
CUDA_CACHE_CLEAR_INTERVAL3Clear CUDA cache every N requests
Clear CUDA cache every N requests
DEVICEautoDevice to use for inference (auto/cuda/mps/cpu)
Device to use for inference (auto/cuda/mps/cpu)
values: auto/cuda/mps/cpu
ENABLE_MEMORY_MONITORINGtrueEnable detailed memory logging
Enable detailed memory logging
values: true/false
EXAGGERATION0.5Emotion intensity/exaggeration level
Emotion intensity/exaggeration level
values: 0.25-2.0
FRONTEND_PORT4321Port for frontend/proxy when using fullstack mode
Port for frontend/proxy when using fullstack mode
HOST0.0.0.0Host to bind the server to
Host to bind the server to
LONG_TEXT_CHUNK_SIZE2500Chunk size for splitting long text
Chunk size for splitting long text
LONG_TEXT_DATA_DIR./data/long_text_jobsDirectory for long text job data and state persistence
Directory for long text job data and state persistence
LONG_TEXT_JOB_RETENTION_DAYS7How many days to keep completed long text jobs
How many days to keep completed long text jobs
LONG_TEXT_MAX_CONCURRENT_JOBS3Maximum number of concurrent long text jobs
Maximum number of concurrent long text jobs
LONG_TEXT_MAX_LENGTH100000Maximum characters allowed for long text TTS
Maximum characters allowed for long text TTS
LONG_TEXT_SILENCE_PADDING_MS200Silence padding between chunks in milliseconds
Silence padding between chunks in milliseconds
MAX_CHUNK_LENGTH280Maximum characters per text chunk
Maximum characters per text chunk
MAX_TOTAL_LENGTH3000Maximum total characters for entire input
Maximum total characters for entire input
MEMORY_CLEANUP_INTERVAL5Cleanup memory every N requests
Cleanup memory every N requests
MODEL_CACHE_DIR./modelsDirectory to cache downloaded models
Directory to cache downloaded models
PORT4123API server port
API server port
TEMPERATURE0.8Sampling temperature
Sampling temperature
values: 0.05-5.0
USE_MULTILINGUAL_MODELtrueEnable multilingual TTS model (23 languages)
Enable multilingual TTS model (23 languages)
values: true/false
VOICE_LIBRARY_DIR./voicesDirectory to store uploaded voice library
Directory to store uploaded voice library
VOICE_SAMPLE_PATH./voice-sample.mp3Path to the voice sample file for cloning
Path to the voice sample file for cloning