Smol TTS keeps getting better! Introducing OuteTTS v0.2 - 500M parameters, multilingual with voice cloning! 🔥 > Multilingual - English, Chinese, Korean & Japanese > Cross platform inference w/ llama.cpp > Zero-shot voice cloning > Trained on 5 Billion audio tokens > Qwen 2.5 0.5B LLM backbone > Trained... via HF GPU grants Model weights on the hub, you can even run this on a Raspberry Pi! Go run, inference now! 🐐show more

Check out the model weights and inference code base here:

llama.cpp compatible GGUFs:

OuteTTS GitHub:

Big Congrats!!! Another SOTA TTS model trained on Emilia after F5-TTS & MaskGCT! Try out:

Just tested it out and the quality is very good. More importantly, the fact that you can generate speaker profiles is awesome! Will test it out some more and add it to my growing list of supported tts engines in my app 🤣

Switch to SkyTab and get $5,000! A modern and sleek POS system with commercial-grade durability. 💪 ✅ $0 upfront costs ✅ Best in-class POS ✅ Local service & 24/7 support ✅ And much more! Make the switch today:

This is improving so fast that I don't want to speak myself anymore. Just use this and get done 🤖

Thank you for your work and for sharing insights! 🙌 Advancements like OuteTTS v0.2 showcase the rapid evolution of AI and its potential to empower global communities. 🚀 The future of #AI is bright, and collaborative innovation is key to unlocking its full potential!

Are you saying you can voice CLONE on a R-Pi? Is that what you're saying????

The font of Japanese characters is wrong, it's for (maybe) Chinese. I hope you'll pay attention and respect to each of them when you are working for multilingual/multicultural things. (like your TTS engine itself does. Brilliant quality✨️)

I tried to run it on HF. average inference time for 200 chars is >1 hour running on CPU. Why is this model so heavy?