How to use Chatterbox: what you can do with it

3 use cases

• Generate speech in a cloned voice with laughs and pauses

  Use Chatterbox-Turbo with a 10 second reference clip and paralinguistic tags for natural sounding lines.

  • REPO resemble-ai/chatterbox
  • Claude Code
  • Codex
  • Cursor
  • Gemini CLI
  • Content
  • Intermediate
  View prompt

  Using resemble-ai/chatterbox, install `pip install chatterbox-tts`, write a script with ChatterboxTurboTTS that reads <my script lines> in the voice from <my 10 second clip>, adding [chuckle] where it fits, and saves one WAV per line.

  Share Email me this

• Narrate the same text in many languages

  Use Chatterbox Multilingual V3 to generate French, Chinese and 20+ other languages from one model.

  • REPO resemble-ai/chatterbox
  • Claude Code
  • Codex
  • Cursor
  • Gemini CLI
  • Content
  • Intermediate
  View prompt

  With resemble-ai/chatterbox, load ChatterboxMultilingualTTS with t3_model="v3" and generate <my paragraph> in French, Spanish and Hindi using language_id, saving each as a WAV.

  Share Email me this

• Run on-device speech on a CPU with Chatterbox-Nano

  Load the 110M Nano model through the Turbo class to generate speech without a GPU.

  • REPO resemble-ai/chatterbox
  • Claude Code
  • Codex
  • Cursor
  • Gemini CLI
  • Content
  • Beginner
  View prompt

  Using resemble-ai/chatterbox, write a CPU-only script that loads ChatterboxTurboTTS with nano=True and device cpu, generates <my sentence>, and times the run.

  Share Email me this

Quick start with Chatterbox

1. Check you have what it needs: Python 3.11 (developed and tested on Debian 11); A CUDA GPU for Turbo and Multilingual; Nano also runs on CPU; A short reference audio clip if you want to clone a voice.
2. Open resemble-ai/chatterbox on GitHub and read the README and license.
3. Install with pip install chatterbox-tts
4. Load ChatterboxTurboTTS.from_pretrained with device set to cuda
5. Call model.generate with your text and audio_prompt_path pointing to your reference clip
6. Save the result with torchaudio as in the README

These steps come from our top use case, Generate speech in a cloned voice with laughs and pauses, which also has a copy-paste agent prompt.

Using the Deduped MCP, get the setup steps for resemble-ai/chatterbox

claude mcp add --transport http deduped https://dedupedai.com/mcp

Best AI repos for Content

About this repo

• Who it is forDevelopers building voice agents, narration or dubbing who want open TTS with voice cloning.
• Why it is hot27k stars, talked about in 2 AI sources we track
• LanguagePython
• Sourcegithub.com/resemble-ai/chatterbox · Back to Hot repos

Requirements

• Python 3.11 (developed and tested on Debian 11)
• A CUDA GPU for Turbo and Multilingual; Nano also runs on CPU
• A short reference audio clip if you want to clone a voice

Related repos

• Trended earlier REPO KittenML/KittenTTS Kitten TTS Kitten-TTS is an open-source family of text-to-speech models, including the 1.7B 1-bit Kitten-TTS 2 with voice cloning and expression control and lightweight 15M to 80M models that run on CPU.
• Trended earlier REPO SYSTRAN/faster-whisper Faster Whisper faster-whisper is a Python reimplementation of OpenAI's Whisper built on the CTranslate2 inference engine. It is up to 4 times faster than openai/whisper at the same accuracy while using less memory, supports 8-bit quantization on CPU and GPU, batched transcription, word-level timestamps, Distil-Whisper checkpoints and a built-in Silero VAD filter.
• Trended earlier REPO Tencent-Hunyuan/AuK AuK AuK is an open-source 1.5B speech model that generates and edits speech from natural-language instructions.