TTS.ai
The most comprehensive open-source AI voice platform.
20+ TTS models, 9 voice cloning models, speech-to-text, AI voice agents, and professional audio tools — all commercially friendly, all open source.
About
TTS.ai provides a unified platform for AI voice generation. We run the best open-source TTS models so you don't have to manage GPU infrastructure. Every model uses commercially-friendly licenses (Apache 2.0, MIT).
Models We Run
| Category | Models |
|---|---|
| TTS | Kokoro, Chatterbox, Sesame CSM, CosyVoice2, Bark, Orpheus, StyleTTS2, GLM-TTS, IndexTTS-2, Spark, Dia, Parler, MeloTTS, Piper, VITS, OpenVoice, Tortoise, GPT-SoVITS, Qwen3-TTS |
| Voice Cloning | Chatterbox, CosyVoice2, GLM-TTS, GPT-SoVITS, IndexTTS-2, OpenVoice, Spark, Tortoise, Qwen3-TTS |
| Speech-to-Text | Faster Whisper, SenseVoice |
| Audio | Demucs (separation), RNNoise (enhancement) |
Links
- Platform: tts.ai
- API Documentation: tts.ai/api
- Python SDK:
pip install ttsai— GitHub - JavaScript SDK:
npm install @ttsainpm/ttsai— GitHub - GitHub: github.com/ttsaigit
Get Started
Try it free at tts.ai or integrate via our API:
from ttsai import TTS
client = TTS(api_key="sk-tts-...")
audio = client.generate("Hello world!", model="kokoro")
audio.save("output.wav")
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support