Implementation of SoundStorm, Efficient Parallel Audio Generation from Google Deepmind, in Pytorch
StreamSpeech is an “All in One” seamless model for offline and simultaneous speech recognition, speech translation and speech synthesis.
