Edge0 releases a free voice cloning model that speaks 23 languages
Edge0 has released Audio8-TTS-Realistic, an open model that turns text into speech and can copy a voice without extra training. It works in 23 languages and is released under the Apache 2.0 open licence, so developers can download and use it. It is best suited to people who build voice apps or tools.
How to try it
Search for Audio8-TTS-Realistic on Hugging Face, where many open-weight models are shared. Download the model files and follow the setup notes to run it on your own computer.
Useful 35%