A DNN Based Speech Enhancement Approach to Noise Robust Acoustic-to-Articulatory Inversion.
Abdolreza Sabzi ShahrebabakiSabato Marco SiniscalchiGiampiero SalviTorbjørn SvendsenPublished in: ISCAS (2021)
Keyphrases
- speech enhancement
- noisy environments
- speech recognition
- noisy speech
- vocal tract
- speech signal
- noise reduction
- signal to noise ratio
- background noise
- speaker verification
- acoustic features
- automatic speech recognition
- speech synthesis
- additive noise
- multiscale
- linear prediction
- wiener filter
- single channel
- image acquisition
- pattern recognition
- image sequences
- information retrieval