Co-occurring keywords · papers mentioning both
Papers
Minimum latency training of sequence transducers for streaming end-to-end speech recognition
Interspeech 2022
Collaborative Training of Acoustic Encoders for Speech Recognition
Interspeech 2021