Investigation into Joint Optimization of Single Channel Speech Enhancement and Acoustic Modeling for Robust ASR.
Tobias MenneRalf SchlüterHermann NeyPublished in: ICASSP (2019)
Keyphrases
- speech enhancement
- noisy environments
- joint optimization
- single channel
- multi channel
- speech signal
- noise reduction
- automatic speech recognition
- speech recognition
- sound source
- background noise
- prior information
- independent component analysis
- signal to noise ratio
- multiscale
- frequency domain
- signal processing
- sparse representation
- wiener filter
- probabilistic model
- vocal tract
- computational complexity