TTS-o-matic local AI text to speech

TTS-o-matic brings you our cutting edge Text-to-speech technology directly to your Unity3D games! Read any string with beautiful and realistic voices powered by AI models!

Any string will be converted to a Unity3D audioclip that can then be used seamlessy in your games: addi realistic voices to your characters, narrate gameplay or use it for accessibility features! TTS-o-matic is fast and easy to integrate!

Features:

⭐ Easy to use: Import the plugin, download your desired voice from a wide selection of pre-trained models (more than 80 voices, 30 languages available!). Get started in less than a minute with the included demo scene!

💻 Offline on-demand generation: no internet connection required! A simple function to select the desired model, and another function to generate the audioclip. All running locally.

⚡ Fast: written in C++ TTS-o-matic brings you fast TTS generation with a low memory footpring, delivering high quality and smooth audio playback.

↔️ Extendable: based on sherpa and compatible with Icefall, custom-trained model using Icefall can be loaded directly in TTS-o-matic.

🔊 Export to .wav: complete and easy to code to export your TTS generated file to a wav file!

Join the Discord server!

Read the Documentation PDF!

Asset uses sherpa-tts under Apache License Version 2.0; see Third-Party notices.txt file in package for details. Asset allows downloads of pre-trained models from the sherpa-tts/icefall project ( https://k2-fsa.github.io/icefall/ ) from a curated selection of models from https://github.com/k2-fsa/sherpa-onnx/releases/tag/tts-models ; Each model comes with its own license. Please refer to the license file included in each download for more information on the license (MODEL_CARD, README.md); see Third-Party notices.txt file in package for details; Pre-trained models for download include languages in English (UK, US), German, Italian, Spanish, Polish, Ukrainian, Russian, French, Korean, Portuguese, and more.