Skip to the content
Know your tech before you buy it. Makers: get listed from $49/mo

Chatterbox TTS

By Chatterbox TTS · github.com

  • Free plan
  • Open source
  • Windows
  • macOS
  • Linux
  • API
  • Self-hosted
8.1 Club score Visit github.com

What Chatterbox TTS is

Chatterbox TTS is a family of open-source text-to-speech models from Resemble AI, available free under the MIT license for research and development. The project lists Turbo, Nano, Multilingual V3, and dedicated single-language models. Turbo supports tags such as [sigh], [gasp], [cough], and [laugh] to prompt vocal reactions; Nano is intended for on-device or CPU inference when memory and latency budgets are tight. The README lists 23 languages for Multilingual and provides pip and source installation instructions. It describes development with Python 3.11 on Debian 11, with CUDA, CPU, and MPS device options. Chatterbox supports voice cloning, batch processing, audio input, voice isolation, and WAV export. Every generated audio file includes Resemble AI’s PerTh watermark. The models are listed for Linux, macOS, and self-hosted use. Resemble says it has stopped offering voice generation as a commercial product, while the models remain open and free for research and development.

Who it suits

Chatterbox TTS may suit developers and researchers looking for open-source speech models. Turbo is aimed at low-latency voice agents, while Nano targets on-device or CPU use with tight memory and latency budgets.

Good to know, and what to watch

  • Free models under the MIT license
  • Multilingual model lists 23 supported languages
  • Supports voice cloning and WAV export
  • Offers CUDA, CPU, and MPS device options
  • Generated audio includes a PerTh watermark
  • Resemble no longer offers voice generation commercially

Chatterbox open-source models

Free

Voice AI Research (Chatterbox)

Free

Free plan

Getting started

  1. Open the Chatterbox repository at https://github.com/resemble-ai/chatterbox
  2. Install the package with pip or install from source
  3. Use Python 3.11 on a supported Linux, macOS, or self-hosted setup
  4. Select Turbo, Nano, Multilingual V3, or a single-language model
  5. Choose CUDA, CPU, or MPS runtime options
  6. Provide audio input and export generated speech as WAV

Limits to check

The models are offered free for research and development under the MIT license. Generated audio carries a PerTh watermark, and processing is batch.

The verdict

Chatterbox TTS offers several open-source model options, including multilingual and single-language choices. Its watermark and research-and-development availability are important considerations for teams planning to use generated audio.

Questions people ask

Is Chatterbox TTS free?

Yes. The listed plan is free under the MIT license for research and development.

Which platforms can run Chatterbox?

The listed platforms are Linux, macOS, and self-hosted environments. Device options include CUDA, CPU, and MPS.

Does Chatterbox support voice cloning?

Yes. Voice cloning is listed as a supported capability.

How many languages does Chatterbox support?

The README lists 23 languages for Chatterbox Multilingual.

What audio format does it export?

Generated audio can be exported as WAV files.

Do generated files contain a watermark?

Yes. Every generated audio file includes Resemble AI’s PerTh watermark.

What the maker publishes

Commercial use
Yes source
Company founding
Resemble says Zohaib Ahmed and Saqib Muhammad started the company in 2019. source
Company headquarters
LinkedIn lists Resemble AI headquarters in Santa Clara, California, with 11–50 employees. source
Company ownership and funding
Crunchbase describes Resemble AI as a private, for-profit company with five funding rounds and founders Zohaib Ahmed and Saqib Muhammad, while funding amounts are obscured. source
Controls
The Python interface exposes generation controls including exaggeration, CFG weight, temperature and random seed. source
Current availability
Resemble says it has stopped offering voice generation as a commercial product and that its voice models remain open and free for research and development. source
Documentation depth
The repository includes model selection tables, Python examples, language identifiers, watermark extraction code, tuning guidance and evaluation links. source
Export formats
WAV source
Expressive speech
Turbo supports paralinguistic tags such as [sigh], [gasp], [cough], and [laugh] to prompt vocal reactions. source
Founded
2019 source
Free plan
Yes source
Hallucination reduction
The V3 description says it reduces unwanted continuation, repetition and off-prompt speech compared with earlier multilingual versions. source
Installation
The README provides a pip installation command and instructions to install from source. source
Installation
The project can be installed with pip as chatterbox-tts or from a cloned repository in editable mode. source
Integration
NVIDIA provides a Chatterbox TTS plugin for ACE for Games with CUDA, Vulkan and Windows D3D12 backends. source
Intended users
The README describes Turbo for low-latency voice agents and Nano for on-device or CPU inference when memory and latency budgets are tight. source
Languages
The README lists 23 supported languages for Chatterbox Multilingual. source
Latest model
The repository identifies Chatterbox Multilingual V3 as its newest general-purpose multilingual model. source
License
The project repository is MIT licensed. source
License
The code and model listing identify the project as MIT licensed, permitting modification and redistribution subject to the license notice. source
Limit: commercial availability
Resemble’s current product notice says voice generation is no longer offered as a commercial product, although the models remain open and free for research and development. source
Limit: comparative performance evidence
An independent review reports that Resemble’s blind test favored Chatterbox, but an Artificial Analysis snapshot ranked Chatterbox below ElevenLabs’ leading model. source
Limit: service reliability and billing
Trustpilot complaints repeatedly mention failed or cut-off audio, slow or absent support, unexpected charges, credit problems and service outages; these reviews concern Resemble’s former hosted service rather than the self-hosted model. source
Limit: watermark removability
An independent review notes that MIT licensing allows modified software, so users can technically remove the watermark by changing the code. source

Checked against the maker's pages on 21 Oct 2026.