Acoustic-to-articulatory inversion for dysarthric speech: Are pre-trained self-supervised representations favorable?
Sarthak Kumar MaharanaKrishna Kamal AdidamShoumik NandiAjitesh SrivastavaPublished in: CoRR (2023)
Keyphrases
- pre trained
- acoustic features
- vocal tract
- speech recognition
- speech signal
- formant frequencies
- speech sounds
- training data
- speech synthesis
- automatic speech recognition
- speaker verification
- multi stream
- visual features
- training examples
- control signals
- audio features
- music information retrieval
- audio visual
- sound source
- hidden markov models
- feature vectors