FSA/FST algorithms, differentiable, with PyTorch compatibility. Automatic speech recognition
Speaker diarization, speake segmentation,
Transcribe audio to text
Convert spoken words into text using voice input