Installation¶
Installation¶
Note
From coqui-tts 0.27.4, PyTorch is not included by default and you need to install it yourself.
🐸TTS is tested on Ubuntu 24.04 with python >= 3.10, < 3.15 and PyTorch 2.2+, but should also work on Mac and Windows.
It is strongly recommended to use uv to install
everything into a virtual environment (otherwise leave out uv from the
commands below).
First install PyTorch, torchaudio, and (only for PyTorch 2.9+) torchcodec
with their official instructions,
choosing the CPU/CUDA/ROCm version as necessary. Or let uv automatically select
the right version for your system:
uv pip install torch torchaudio torchcodec --torch-backend=auto
If you are only interested in synthesizing speech with the pretrained 🐸TTS models, installing from PyPI is the easiest option.
uv pip install coqui-tts
If you plan to code or train models, clone 🐸TTS and install it locally.
git clone https://github.com/idiap/coqui-ai-TTS
cd coqui-ai-TTS
uv pip install -e .
Optional dependencies¶
The following extras allow the installation of optional dependencies:
Name |
Description |
|---|---|
|
All optional dependencies |
|
Dependencies only used in notebooks |
|
Dependencies to run the TTS server |
|
Bangla G2P |
|
Japanese G2P |
|
Korean G2P |
|
Chinese G2P |
|
All language-specific dependencies |
You can install extras with one of the following commands:
uv pip install coqui-tts[server,ja]
uv pip install -e .[server,ja]
Pytorch extras¶
There are also the following convenience extras to automatically install the
PyTorch dependencies. Note that the CPU/CUDA selection only works with uv and
when installing Coqui from source. With other package managers or when installing
coqui-tts from PyPI, the PyTorch dependencies will be installed from PyPI.
Name |
Description |
|---|---|
|
Install |
|
Install |
|
Install |
|
Install |
Platforms¶
If you are on Ubuntu (Debian), you can also run the following commands for installation.
make system-deps
make install