Chatterbox
Chatterbox is a family of open-source text-to-speech models from Resemble AI, released under MIT for both the code and the weights. It clones a voice from a reference clip of about ten seconds, its multilingual model covers 23 languages, and its Turbo and Nano models accept tags such as [laugh]. It is the voice cloner on this site whose licence permits commercial use without a separate agreement.
What does Chatterbox do?
Chatterbox is a family of text-to-speech models published by Resemble AI. The README describes five models: Chatterbox-Turbo (350 million parameters, English), Chatterbox-Nano (110 million, English), Chatterbox-Multilingual V3 (500 million, 23 languages), a set of six single-language finetunes at 500 million each, and the original 500 million English model.
Voice cloning is built in. You pass a reference clip, and the README's examples use a ten-second file, so no training step is needed. The multilingual model lists Arabic, Danish, German, Greek, English, Spanish, Finnish, French, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Dutch, Norwegian, Polish, Portuguese, Russian, Swedish, Swahili, Turkish and Chinese.
The Turbo and Nano models accept paralinguistic tags written into the text, such as [cough], [laugh] and [chuckle]. Every generated file carries Resemble's Perth watermark, and the README includes a script to detect it.
Key facts
- Five models in the README: Chatterbox-Turbo (350 million parameters, English), Chatterbox-Nano (110 million, English), Chatterbox-Multilingual V3 (500 million, 23 languages), six single-language finetunes at 500 million each, and the original 500 million English model.
- Licence: MIT on the code (LICENSE file) and MIT on the weights (Hugging Face model card, ResembleAI/chatterbox).
- Voice cloning is zero-shot from a reference clip; the README's example uses a ten-second WAV file.
- The 23 languages of the multilingual model include Arabic, German, English, Spanish, French, Hindi, Japanese, Korean, Portuguese, Russian, Turkish and Chinese.
- Turbo and Nano accept paralinguistic tags written into the text, such as [cough], [laugh] and [chuckle].
- Every generated file carries Resemble's Perth watermark, and the README ships a detection script.
- The README describes Nano as "running 3x faster than realtime on 8 CPU cores" and as the model to pick "when memory and latency budgets are tightest".
- Developed and tested on Python 3.11 on Debian 11; devices in the examples are cuda, cpu and mps.
What does it replace, and where does it fall short?
Chatterbox replaces ElevenLabs for voice generation and voice cloning where the licence has to permit commercial use. The code is MIT and the weights on Hugging Face are MIT, which is the reason it sits at the top of the ElevenLabs list on this site.
Where it falls short: the README does not mention streaming, and it does not state a sample rate. ElevenLabs' own claim is 70+ languages against Chatterbox's 23. Voice quality is not compared here, because nobody has measured both on the same test. The demo page is linked above so you can listen with your own text.
How does Chatterbox run?
pip install chatterbox-tts, or clone the repository and install in editable mode. It is a Python library, so serving it over HTTP means writing a small wrapper. The examples pass a device of cuda, cpu or mps. The README says Nano "also runs on CPU" and describes it "running 3x faster than realtime on 8 CPU cores". Development and testing were on Python 3.11 on Debian 11; Windows and macOS are not named.
Resemble AI sells a hosted service, and the README points to it "if you like the model but need to scale or tune it".
Who is Chatterbox for?
A team that wants voice cloning it can ship in a product, with a licence it does not have to negotiate. A team that needs many languages beyond the 23 listed, or a streaming API without writing one, should read the other entries on the ElevenLabs page.
What limits does the README state?
The reference clip should match the language tag, or the output "may inherit the accent of the reference clip's language". A higher exaggeration setting "tends to speed up speech". The latest tagged GitHub release is from June 2025; the README's own "latest release" heading refers to the Multilingual V3 model, which has no tag.
Questions people ask
Is Chatterbox open source?
Yes. The code is under the MIT licence, and the model weights on Hugging Face are also MIT. That puts both halves on the OSI approved list, which is why this site labels it open source rather than open weights. The training prompts are described only as sourced from freely available data on the internet.
Can Chatterbox be used commercially?
Yes, under the MIT licence on both code and weights, with no separate agreement required. Resemble AI also sells a hosted service, and the README points to it for teams that need to scale or tune for higher accuracy, but nothing in the licence requires using it.
Does Chatterbox run offline, and on what hardware?
Yes. Once the model is downloaded it runs locally with no account or network call. The examples pass a device of cuda, cpu or mps, and the README says Nano also runs on a CPU, at three times real time on 8 cores. Development and testing were on Debian 11; Windows and macOS are not named.
How does Chatterbox compare with ElevenLabs?
Chatterbox gives you cloning and 23 languages under a licence you can ship with, on your own hardware, with no per-character bill. ElevenLabs claims 70+ languages and 5,000+ voices and needs no operations work. The open-source alternatives to ElevenLabs page sets Chatterbox beside Kokoro, Piper, Fish Speech and F5-TTS by need.
Which Chatterbox model should I start with?
The README recommends Nano when memory and latency are tightest, Turbo for English with the expressive tags, and Multilingual V3 for the 23 languages. Match the reference clip's language to the language tag: the README warns that a mismatch makes the output inherit the clip's accent.
Sources
- Chatterbox README and LICENSE: github.com/resemble-ai/chatterbox, read 2026-09-16.
- Weights licence on the model card: huggingface.co/ResembleAI/chatterbox, read 2026-09-16.
Compared with the others
On the open-source alternatives to ElevenLabs page, Chatterbox is the pick for voice cloning you can ship. MIT code and MIT weights, cloning from a reference clip of about ten seconds, and a multilingual model covering 23 languages. The one cloner here whose licence needs no negotiation.
Also on that page: Kokoro FastAPI for a speech api on your own server, Piper for small devices and offline assistants, Fish Speech for expressive cloning, non-commercial.
More text to speech and voice cloning
An AGPL-3.0 desktop workflow engine for voice cloning, voice design, video dubbing, dictation and audiobooks, running 16 speech engines and 10 transcription engines locally, with a local API and an MCP server for agents.
A multilingual text-to-speech system with voice cloning from a 10 to 30 second clip and inline emotion tags, released with its weights under a research licence that requires a separate agreement for commercial use.
A research text-to-speech model with zero-shot voice cloning whose code is MIT but whose pretrained weights are non-commercial, because of the dataset they were trained on.
A fast, local neural text-to-speech engine maintained by the Open Home Foundation, used by Home Assistant and the NVDA screen reader, under GPL-3.0.
A Docker image that serves the 82-million-parameter Kokoro model through an OpenAI-compatible speech endpoint, with streaming, voice mixing and nine languages, all under Apache-2.0.
Tencent Hunyuan's 1.5-billion-parameter speech foundation model under MIT for code and weights, covering zero-shot text to speech, speech editing, enhancement and separation through one natural-language instruction interface.
Added September 16, 2026. Every claim above comes from the project's README, LICENSE or model card, read on September 16, 2026, or from the GitHub API on the date shown in the panel. Found an error? Write to reveneau@licheo.com and it is fixed in the next weekly pass. Repository: github.com/resemble-ai/chatterbox.
Want Chatterbox running on your own servers?
Reveneau sets it up on your hardware, connects it to your product, and builds the software around it.