Investigation of self-supervised pre-trained models for classification of voice quality from speech and neck surface accelerometer signals

Sudarsana Reddy Kadiri, Farhad Javanmardi, Paavo Alku

Abstract: Highlights•Layer-wise analysis of self-supervised pre-trained models for voice quality classification.•Systematic comparison of conventional features and pre-trained model embeddings.•Deriving glottal source waveforms with two methods.•Studying voice quality classification with features from glottal source waveforms.