Skip to content
ML2
Creative & MediaTool#Audio#GitHub repo#Local / BYO model#Open source

Chatterbox TTS

Added New⧉ Copyable prompt below

Our verdict

A solid open option when you want narration or voice output without per-character API fees. Turbo is the English model aimed at low-latency agents and supports tags like [laugh]; the multilingual model covers 23+ languages. Voice cloning needs a reference clip: only clone voices you have permission to use.

Who it's for
Builders adding narration or voice to videos, apps and agents
Leverage
Local voice generation without per-use fees
Effort
Medium
Date added

What it does

Chatterbox is an MIT-licensed family of text-to-speech models by Resemble AI: Chatterbox-Turbo (350M, English, paralinguistic tags), Chatterbox-Nano (110M, CPU-friendly), and Chatterbox-Multilingual V3 (500M, 23+ languages), plus single-language finetunes. It installs with pip install chatterbox-tts and runs from a few lines of Python.

How builders use it

  • pip install chatterbox-tts, then load ChatterboxTurboTTS and generate from a short reference clip.
  • Use Nano with device="cpu" when you have no GPU.
  • Add tags like [chuckle] in the script for more natural narration.
  • Try the hosted Hugging Face demos before installing locally.

What to expect

Voiceovers and agent speech generated on your own hardware.

Copy this: Minimal Python

import torchaudio as ta
from chatterbox.tts_turbo import ChatterboxTurboTTS

model = ChatterboxTurboTTS.from_pretrained(device="cuda")  # or device="cpu", nano=True
wav = model.generate("Welcome back [chuckle], here's what's new this week.", audio_prompt_path="my_voice_10s.wav")
ta.save("intro.wav", wav, model.sr)

Related discoveries

  • ToolCreative & Media

    ComfyUI

    Open-source node-graph engine for image, video and audio generation that runs locally and exposes an API.

    NewEffort: High#Image#Video

  • ToolCreative & Media

    Invoke (InvokeAI)

    An open-source, locally run image studio with a canvas, inpainting and node workflows for open models.

    NewEffort: Medium#Image#GitHub repo

  • ToolCreative & Media

    LTX-2: open audio and video generation

    Lightricks' open-weights model for video with synchronized audio, runnable locally or inside ComfyUI.

    NewEffort: High#Video#Audio

  • ToolCoding & Shipping Agents

    Aider

    Open-source AI pair programming in your terminal, with git commits built in.

    Effort: Low#GitHub repo#Local / BYO model

Want more like this?

Practical discoveries, without the noise.

Get the newsletterMore in Creative & Media