Paaploss: A Phonetic-Aligned Acoustic Parameter Loss for Speech Enhancement.
Muqiao YangJoseph KonanDavid BickYunyang ZengShuo HanAnurag KumarShinji WatanabeBhiksha RajPublished in: ICASSP (2023)
Keyphrases
- speech enhancement
- noisy speech
- speech sounds
- speech recognition
- vocal tract
- speech signal
- noisy environments
- noise reduction
- signal to noise ratio
- single channel
- emotional speech
- automatic speech recognition
- linear prediction
- hidden markov models
- background noise
- additive noise
- smoothing algorithm
- multiresolution
- speaker identification
- image segmentation
- image coding
- acoustic features
- sound source
- image processing
- multi channel