OpenL3 embeddings outperform VGGish and Wav2Vec2.0 for Parkinson's disease speech classification on the NeuroVoz dataset, and Wav2Vec2.0 shows a male-favoring gender bias in the diadochokinesis task.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.AS 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Evaluating the Effectiveness of Pre-Trained Audio Embeddings for Classification of Parkinson's Disease Speech Data
OpenL3 embeddings outperform VGGish and Wav2Vec2.0 for Parkinson's disease speech classification on the NeuroVoz dataset, and Wav2Vec2.0 shows a male-favoring gender bias in the diadochokinesis task.