Skip to main content

Free plan cut on 6 Nov 2026. Was: Open source: run the models yourself with pip and Python, no usage limit. Now: Open source: MIT-licensed text-to-speech models, archived. See it on our free plan changes

Chatterbox

Open-source text-to-speech models from Resemble AI with voice cloning, run on your own machine with Python.

Try free
Rate it

Is Chatterbox free?

Yes, Chatterbox has a free plan. What you get: MIT-licensed text-to-speech models, archived. To get it: install Python 3.11, then run pip install chatterbox-tts; then load a model in Python and call generate with your text; then add a reference clip for voice cloning, then save the audio. Verified on github.com.

Free plan at a glance

Includes
  • MIT-licensed text-to-speech models, archived

Open source under the MIT license and run on your own machine. The README says it is retired research: no support, updates or paid tier.

Duration
Free forever
Watermark
Yes, on free outputs (a watermark or the maker's branding)

How to get it free

Self-host
  1. Install Python 3.11, then run pip install chatterbox-tts
  2. Load a model in Python and call generate with your text
  3. Add a reference clip for voice cloning, then save the audio

Good to know

  • Every generated audio file carries a Perth watermark
  • Retired: the README says "As of August 2026, Chatterbox is retired research and is no longer maintained."
  • The README conflicts on paid options: "with no support, updates, or paid tier" in one place, "check out our competitively priced TTS service" in another
Get the free plan

github.com, Checked Source

Overview

Chatterbox is a family of open-source text-to-speech models from Resemble AI. You install it with pip and run it from Python; the multilingual model supports voice cloning across many languages.

Key features

  • Voice cloning from a short reference audio clip
  • Multilingual model covering languages such as Arabic, Hindi, Japanese and Chinese
  • Turbo model for low-latency English voice agents, with tags such as [laugh]
  • Nano model that can run on a CPU
  • Sample scripts and Gradio apps in the repository

Best for

People
Developers adding speech to apps or voice agents, Creators who want to clone a voice from a short clip, Teams that need multilingual speech they can run themselves

Limitations

  • Needs a Python install and code to use; the README describes no hosted app
  • Every generated audio file carries an imperceptible Perth watermark
  • Tested on Python 3.11, on Debian 11, with pinned dependencies
  • Retired research, no longer maintained as of August 2026: no support or updates

FAQ

Is Chatterbox free?
Yes. Chatterbox has a free plan. That's from github.com, checked on November 17, 2026.
What does the Chatterbox free plan include?
What's free: MIT-licensed text-to-speech models, archived. Open source under the MIT license and run on your own machine. The README says it is retired research: no support, updates or paid tier. We read this on github.com/resemble-ai/chatterbox on November 17, 2026.
How do I get Chatterbox for free?
Install Python 3.11, then run pip install chatterbox-tts. Then load a model in Python and call generate with your text. Finally, add a reference clip for voice cloning, then save the audio. Start at github.com/resemble-ai/chatterbox. We last checked these steps on November 16, 2026.
Do I need an account to use Chatterbox for free?
Not for the free option: the free way to use Chatterbox is to host it yourself. To start, install Python 3.11, then run pip install chatterbox-tts (checked November 16, 2026).
What's the catch with Chatterbox's free plan?
Two things to know. Every generated audio file carries a Perth watermark. Retired: the README says "As of August 2026, Chatterbox is retired research and is no longer maintained.". We last checked this on November 16, 2026.
What are free alternatives to Chatterbox?
NaturalReader, Speechma and Kitten-TTS all have free plans. We checked their free offers between November 16, 2026 and November 17, 2026.
Which languages does Chatterbox support?
The general-purpose Chatterbox Multilingual model supports the languages listed in the README, including Arabic, Hindi, Japanese and Chinese.Source: github.com/resemble-ai/chatterbox
Do generated files carry a watermark?
Yes. The README says every audio file includes Resemble AI's Perth watermark, which is imperceptible and can be detected with a script.Source: github.com/resemble-ai/chatterbox
Is there a paid version?
The README is not consistent. One paragraph says the code stays public "with no support, updates, or paid tier"; another still suggests Resemble AI's "competitively priced TTS service". We list only the free, MIT-licensed code.Source: github.com/resemble-ai/chatterbox

Free alternatives

More free AI Audio & Voice tools

All free AI Audio & Voice tools

Sources

Free offer
Verified on github.com/resemble-ai/chatterbox
…code stays public as an archive under the MIT license, with no support, updates, or paid…
Page content

Checked against

How we check free offers

Added by our team. We check that a free option exists and record the date; signing up is between you and Chatterbox, so read their privacy policy and terms first.

Is this your tool? Claim it · Promote it