Coqui TTS
A neural text-to-speech toolkit you can fine-tune on your own voices, frozen but still running.
What it is
Coqui TTS is a Python library for turning text into speech with neural models. It bundles a long list of pretrained voices, a training pipeline for your own, and voice cloning tools, so you can go from a text file to spoken audio without writing model code yourself.
It grew out of Mozilla TTS and was developed by Coqui, a company that also ran a hosted service. The company shut down in early 2024, so the original repository is effectively frozen: the code and the PyPI package still work, but new development happens in community forks rather than upstream.
It is a reasonable choice when you need an open, self-hosted engine with many voices and the ability to fine-tune, and you accept that it will not receive upstream fixes. If you need something actively maintained for production, evaluate the community forks or a newer engine before committing.
Why we recommend it
Voice cloning without a cloud subscription
You can fine-tune a model on a few minutes of a speaker and generate new lines in that voice, entirely on your own machine. Hosted TTS products charge for this and restrict how you use it.
Many languages and speakers in one package
Pretrained models cover a long list of languages and the same API switches between them. For multilingual projects that avoids stitching several vendors together.
It is a training framework, not just a demo
The repository ships training recipes, data loaders, vocoders and evaluation tools. If the provided voices are not good enough, you have everything needed to build your own.
The code still installs even though the company is gone
The repository and the PyPI package remain available. Treat it as a stable, frozen tool: expect what exists today to keep working, and expect nothing new.
How it scores
These are our editorial ratings, not benchmark results. We score on how the project behaves in practice — how long it takes to get running, how good the documentation is, how responsive the maintainers are, and how well it holds up over time. The licence and star count above are read from GitHub and are a snapshot.
Pros and cons
Strengths
- Large library of pretrained voices and vocoders
- Voice cloning and multi-speaker fine-tuning supported out of the box
- Both a Python API and a command line tool over the same models
- Runs offline with no API key and no per-character billing
- MPL 2.0 licence, which is permissive enough for most commercial use
Trade-offs
- The original company has closed and the repository is largely unmaintained — do not expect upstream bug fixes
- Installation is picky about Python and dependency versions, and CUDA setup takes real effort
- Generating audio is slow on CPU; a GPU is strongly recommended
- Some documentation and pretrained model links have rotted since the project stalled
How to get it
One-line install
Prefer the terminal? Copy a command and you are done.
pip install TTS
Source and documentation
At a glance
| Category | Speech AI |
|---|---|
| Licence | MPL 2.0 |
| Written in | Python |
| GitHub stars | 45,999 |
| Repository | coqui-ai/TTS |
Licence and star count were read from the GitHub API when this page was generated and are a snapshot. Check the repository for the current figure and the authoritative licence text.
Other Speech AI tools
We are not affiliated with Coqui TTS or its maintainers. We list it because it is genuinely useful and genuinely free. If you plan to build a commercial product on top of it, read the licence on the repository first.