#2424th of 38 in VoiceFor voice, speech and music
kittentts
Text-to-speech library with voice cloning and CPU-capable models
Live demo ↗
(opens in a new tab)Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 16.2k
- License
- Apache-2.0
- Last commit
- Oct 2026
- Last release
- Feb 2026
- Language
- Python
13,236 on 31 Mar 202616,212 on 20 Oct 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Python library for text-to-speech. The main model, KittenTTS 2 (1.7B parameters), has 47 built-in voices, 20 languages, voice cloning from 5-30 seconds of audio, and emotion tags. It also ships 15M-80M parameter ONNX models that run on CPU. Backends include a llama.cpp fork, MLX on Apple silicon, and vLLM on NVIDIA GPUs.
Who it is for: Developers adding self-hosted speech synthesis and voice cloning to apps
Strengths
- Voice cloning from a 5-30 second clip, no fine-tuning
- Small ONNX models (25-80 MB) run on CPU without a GPU
- Backends for CPU (GGUF/C++), Apple MLX and vLLM CUDA
- Docs describe a Docker setup with an OpenAI-compatible speech API
Weaknesses
- KittenTTS 2 weights use the Stellon Labs Community License, not Apache-2.0
- Streaming is chunk-level; first audio arrives after a real delay
- Expression control and the C++ CPU backend are marked beta
- KittenTTS 2 needs about 8 GB GPU memory or 6 GB RAM
What it needs
- RAM ≥ 6 GB
- GPU optional
- Needs Python 3.9+, ONNX runtime, Whisper (auto-transcribes cloning reference)
- Models: KittenTTS 2 (1.7B), kitten-tts-mini 80M ONNX, kitten-tts-micro 40M ONNX, kitten-tts-nano 15M ONNX, GGUF (C++ backend)
Also in Voice
See all 38| Rank | Project | Score |
|---|---|---|
| 1 | voicestudioDesktop app for voice cloning, dubbing and dictation with local speech models | 83 out of 100 |
| 2 | whisper.cppC/C++ port of OpenAI Whisper for local speech-to-text transcription | 77 out of 100 |
| 3 | VoiceboxLocal voice studio for cloning, TTS, dictation and agent speech | 71 out of 100 |
| #4 | buzzOffline desktop app for transcribing and translating audio with Whisper | 69 out of 100 |
| #5 | meetilyDesktop app that records, transcribes and summarizes meetings locally | 68 out of 100 |