4 papers · 1 filter
Identifying and typifying demographic unfairness in phoneme-level embeddings of self-supervised speech recognition models
Felix Herron, Solange Rossato, Alexandre Allauzen +1
Modern automatic speech recognition (ASR) systems have been observed to function better for certain speaker groups (SGs) than others, despite recent gains in overall performance. O…
Where Do Self-Supervised Speech Models Become Unfair?
Felix Herron, Maja Hjuler, Solange Rossato +2
Speech encoder models are known to model members of some speaker groups (SGs) better than others. However, there has been little work in establishing why this occurs on a technolog…
Polynomial Mixing for Efficient Self-supervised Speech Encoders
Eva Feillet, Ryan Whetten, David Picard +1
State-of-the-art speech-to-text models typically employ Transformer-based encoders that model token dependencies via self-attention mechanisms. However, the quadratic complexity of…
Interference Matrix: Quantifying Cross-Lingual Interference in Transformer Encoders
Belen Alastruey, João Maria Janeiro, Alexandre Allauzen +3
In this paper, we present a comprehensive study of language interference in encoder-only Transformer models across 83 languages. We construct an interference matrix by training and…