A CNN-based approach to identification of degradations in speech signals.
Yuki SaishuAmir Hossein PoorjamMads Græsbøll ChristensenPublished in: EURASIP J. Audio Speech Music. Process. (2021)
Keyphrases
- speech signal
- speech recognition
- automatic speech recognition
- automatic speech recognition systems
- adaptive filtering
- vocal tract
- spectral analysis
- noisy environments
- sound signals
- non stationary
- additive noise
- linear prediction
- speech enhancement
- background noise
- speaker identification
- blind source separation
- speech sounds
- speech quality
- cellular neural networks
- image segmentation
- hidden markov models
- image quality
- sound source
- multi modal
- multiscale
- machine learning
- noise reduction