Feature Compensation for Speech Recognition in Severely Adverse Environments Due to Background Noise and Channel Distortion.
Wooil KimJohn H. L. HansenPublished in: INTERSPEECH (2011)
Keyphrases
- speech recognition
- background noise
- speech signal
- noisy environments
- noisy speech
- cepstral coefficients
- hidden markov models
- language model
- speech synthesis
- automatic speech recognition
- speech recognition systems
- pattern recognition
- speech enhancement
- speech recognizer
- adaptive filtering
- low contrast
- speaker identification
- speech recognition technology
- feature vectors
- feature set
- speaker dependent
- speaker diarization
- linear prediction
- multi channel
- multiscale
- sound source
- machine learning
- vocal tract
- mel frequency cepstral coefficients
- dynamic environments
- speaker independent
- multi modal
- image analysis
- spectral subtraction
- image processing