title: Chatterbox-Multilingual-TTS emoji: 🌎 colorFrom: indigo colorTo: blue sdk: gradio sdk_version: 5.29.0 app_file: app.py pinned: false short_description: Chatterbox TTS supporting 23 languages

Chatterbox TTS

A high-quality Multilingual Text-to-Speech (TTS) library with voice cloning capabilities, supporting 23 languages.

Features

• 🗣️ Support for 23 languages
• 🎙️ Voice cloning from reference audio
• 🚀 Fast inference with PyTorch
• 🎯 High-quality, natural-sounding speech
• 🔌 Simple Python API

Installation

pip install chatterbox-tts

Note: This package requires PyTorch. For optimal performance, install the appropriate version for your system from PyTorch's official website.

Quick Start

from chatterbox import ChatterboxMultilingualTTS, SUPPORTED_LANGUAGES
import torch

# Check available languages
print("Supported languages:", list(SUPPORTED_LANGUAGES.keys()))

# Initialize the TTS model
device = "cuda" if torch.cuda.is_available() else "cpu"
tts = ChatterboxMultilingualTTS.from_local(device=device)

# Generate speech from text
text = "Hello, this is a test of the Chatterbox TTS system."
language = "en"  # English

# Generate audio
sample_rate, audio = tts.synthesize(
    text=text,
    language=language,
    temperature=0.8,
    cfg_scale=0.5,
    exaggeration=0.5
)

# Save to file
import soundfile as sf
sf.write("output.wav", audio, sample_rate)

Advanced Usage

Voice Cloning

# Generate speech in the style of a reference audio
reference_audio = "path/to/reference_audio.wav"
sample_rate, audio = tts.synthesize(
    text="This will sound like the reference speaker.",
    language="en",
    reference_audio=reference_audio,
    temperature=0.7,
    cfg_scale=0.7
)

Available Languages

Chatterbox TTS supports the following languages:

• English (en)
• Spanish (es)
• French (fr)
• German (de)
• Italian (it)
• Portuguese (pt)
• Polish (pl)
• Russian (ru)
• Japanese (ja)
• Korean (ko)
• Chinese (zh)
• And more... (23 languages total)

License

MIT