Speaking with a KN95 Face Mask: ASR Performance and Speaker Compensation.
Sarah E. GutzHannah P. RoweJordan R. GreenPublished in: Interspeech (2021)
Keyphrases
- automatic speech recognition
- speech recognition
- speech signal
- human faces
- speaker identification
- broadcast news
- image processing
- spontaneous speech
- speaker verification
- noisy environments
- facial images
- facial features
- face images
- facial expressions
- hidden markov models
- multiscale
- case study
- acoustic features
- genetic algorithm
- conversational speech