Fusing CQCC spectral features with Wav2Vec2.0 embeddings via cross-attention lowers average equal error rate from 10.87% to 6.80% across four speech deepfake benchmarks.
Combining Spectral and Self-Supervised Features for Low Resource Speech Recognition and Translation
1 Pith paper cite this work. Polarity classification is still indexing.
abstract
Self-Supervised Learning (SSL) models have been successfully applied in various deep learning-based speech tasks, particularly those with a limited amount of data. However, the quality of SSL representations depends highly on the relatedness between the SSL training domain(s) and the target data domain. On the contrary, spectral feature (SF) extractors such as log Mel-filterbanks are hand-crafted non-learnable components, and could be more robust to domain shifts. The present work examines the assumption that combining non-learnable SF extractors to SSL models is an effective approach to low resource speech tasks. We propose a learnable and interpretable framework to combine SF and SSL representations. The proposed framework outperforms significantly both baseline and SSL models on Automatic Speech Recognition (ASR) and Speech Translation (ST) tasks on three low resource datasets. We additionally design a mixture of experts based combination model. This last model reveals that the relative contribution of SSL models over conventional SF extractors is very small in case of domain mismatch between SSL training set and the target language data.
citation-role summary
citation-polarity summary
fields
cs.SD 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Two Views, One Truth: Spectral and Self-Supervised Features Fusion for Robust Speech Deepfake Detection
Fusing CQCC spectral features with Wav2Vec2.0 embeddings via cross-attention lowers average equal error rate from 10.87% to 6.80% across four speech deepfake benchmarks.