On monoaural speech enhancement for automatic recognition of real noisy speech using mixture invariant training.
Jisi ZhangCatalin ZorilaRama DoddipatlaJon BarkerPublished in: CoRR (2022)
Keyphrases
- speech enhancement
- automatic recognition
- noisy speech
- noisy environments
- noise reduction
- signal to noise ratio
- single channel
- speech signal
- linear prediction
- vocal tract
- speech recognition
- background noise
- hidden markov models
- wiener filter
- additive noise
- smoothing algorithm
- edge preserving
- multi channel
- speech synthesis
- maximum likelihood
- signal processing