Listen, Adapt, Better WER: Source-free Single-utterance Test-time Adaptation for Automatic Speech Recognition.

Guan-Ting Lin, Shang-Wen Li, Hung-yi Lee

Browse the full Interspeech paper archive.