Kitten TTS Installation Guides

Choose your platform and get Kitten TTS running in minutes. All guides cover the latest v0.8.1 release from KittenML.

Platforms

pip Command

80MB

Max Model Size

CPU

No GPU Needed

🪟

Windows 10 / 11

Complete setup guide for Windows. Covers Python environment, pip wheel installation from GitHub releases, optional CUDA GPU acceleration, and troubleshooting common issues.

macOS (Apple Silicon)

Install Kitten TTS on M1/M2/M3/M4 Macs. Runs natively on Apple Silicon with excellent CPU performance via the ONNX inference engine.

🐧

Linux (Ubuntu / Debian / Fedora)

Server and desktop Linux setup. Includes systemd service configuration for production deployments and optional CUDA GPU acceleration.

🐳

Docker Container

Containerize Kitten TTS for production. Dockerfile, docker-compose with GPU passthrough, and multi-stage build optimization.

Quick Start (All Platforms):

pip install soundfile
pip install https://github.com/KittenML/KittenTTS/releases/download/0.8.1/kittentts-0.8.1-py3-none-any.whl

Then: from kittentts import KittenTTS and you are ready to generate speech with 8 built-in voices.

No GPU? No problem. Kitten TTS is built on ONNX and optimized for CPU inference. All four model variants (nano 15M, micro 40M, mini 80M) run in real-time on modern CPUs. GPU is optional and only needed for maximum throughput scenarios.

🚀 Ready to Create AI Videos with Kitten TTS?

Turn scripts into professional videos automatically. Combine Kitten TTS with JNTV for end-to-end AI video production.

Try JNTV Workflow →