Mycroft AI, creators of an open-source speech technology platform, has released the Mimic 3 text-to-speech (TTS) engine with more than two dozen languages and 100 voice sets. It can run completely offline on low-end systems.
With Mimic 3, user data isn’t stored on a cloud and will never be sold, it runs completely offline, is compatible with Raspberry Pi 4 or greater using Mycroft’s TTS Plugin, runs on Mycroft.AI, Linux, or as a Docker container and includes its web server. Users can switch voices, control rate and variability of speech.
Mimic 3 converts all input text to phonemes and recognizes and matches the distinct nuances of how an individual speaks.
