Phoneme class based feature adaptation for mismatch acoustic modeling and recognition of distant noisy speech.
Seçkin UluskanAbhijeet SangwanJohn H. L. HansenPublished in: Int. J. Speech Technol. (2017)
Keyphrases
- noisy speech
- speech recognition
- automatic speech recognition systems
- hidden markov models
- speech signal
- speech synthesis
- background noise
- automatic speech recognition
- noisy environments
- speech enhancement
- language model
- pattern recognition
- signal to noise ratio
- speech recognition systems
- speaker independent
- machine learning
- speech sounds
- feature set
- object recognition
- spectral subtraction
- feature selection
- signal processing
- linear prediction
- adaptive filtering
- speaker identification
- feature vectors