The field of speech processing has undergone a transformative shift with the advent of deep
learning. The use of multiple processing layers has enabled the creation of models capable …

Despite being trained on massive and diverse datasets, speech self-supervised encoders
are generally used for downstream purposes as mere frozen feature extractors or model …

Transfer learning from large multilingual pretrained models, like XLSR, has become the new
paradigm for Automatic Speech Recognition (ASR). Considering their ever-increasing size …

The number of end-to-end speech recognition models grows every year. These models are
often adapted to new domains or languages resulting in a proliferation of expert systems that …

The expansion of speech models emphasizes the importance of parameter efficiency in
practical automatic speech recognition (ASR) systems. Parameter sharing, which reuses the …

Recently, Whisper, an extra-large end-to-end ASR model, was released and demonstrated
powerful capability in multilingual ASR. Fine-tuning Whisper, however, can lead to …