Releases: pensieves/sonorus
Releases · pensieves/sonorus
sonorus 0.1.1 Release with praat-parselmouth and pykaldi specific modules
Added support for praat-parselmouth and pykaldi specific task modules and code such as gender voice conversion and phoneme segmentation respectively.
Initial Release
Implementation of google cloud's and Facebook's Wav2Vec2 Hugging Face Speech-to-Text (STT) models with streaming transcription, Voice Activity Detection (VAD), kenlm and Fairseq language model based text decoding.