On the use of channel-attentive MFCC for robust recognition of partially corrupted speech.
Hoon-Young ChoYung-Hwan OhPublished in: IEEE Signal Process. Lett. (2004)
Keyphrases
- robust recognition
- speech recognition
- speech signal
- speaker identification
- speaker recognition
- mel frequency cepstral coefficients
- automatic speech recognition
- partial occlusion
- multi channel
- speech recognition systems
- cepstral coefficients
- language model
- speaker diarization
- hidden markov models
- feature extraction
- noisy environments
- audio features
- acoustic features
- feature vectors
- linear predictive
- broadcast news
- speaker verification
- state space
- search space
- face recognition
- computer vision