Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/KittenML/KittenTTS/llms.txt

Use this file to discover all available pages before exploring further.

KittenTTS requires Python 3.8 or higher and runs on Linux, macOS, and Windows. Models run on CPU only — no GPU required.

​
System requirements

  • Python 3.8+
  • pip
  • 25–80 MB disk space per model
  • Linux, macOS, or Windows

​
Install KittenTTS

1

Create a virtual environment

Using a virtual environment keeps your project dependencies isolated.
python -m venv .venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate
2

Install KittenTTS

Install directly from the GitHub release wheel:
pip install https://github.com/KittenML/KittenTTS/releases/download/0.8.1/kittentts-0.8.1-py3-none-any.whl
This installs KittenTTS and all required dependencies: onnxruntime, soundfile, numpy, spacy, misaki, espeakng_loader, num2words, and huggingface_hub.
3

Verify the installation

from kittentts import KittenTTS
print("KittenTTS installed successfully")
Models are downloaded automatically from Hugging Face the first time you load them. They are cached locally for subsequent runs.

​
Troubleshooting

KittenTTS uses espeakng_loader to handle phoneme conversion. On some Linux systems, you may need to install the eSpeak NG system library:
# Debian/Ubuntu
sudo apt-get install espeak-ng

# Fedora/RHEL
sudo dnf install espeak-ng

# macOS
brew install espeak-ng
On Windows and macOS, espeakng_loader typically bundles the required binaries, so no additional steps are needed.
If you see errors related to spaCy language models, install the required English model:
python -m spacy download en_core_web_sm
If pip cannot reach the GitHub release URL, try downloading the wheel manually and installing from the local file:
pip install kittentts-0.8.1-py3-none-any.whl
KittenTTS requires Python 3.8 or higher. Check your version:
python --version
If you are on an older version, install a supported Python release from python.org.

Build docs developers (and LLMs) love

Get started for freeTalk to us