Linear histogram equalization in the acoustic feature domain for speech recognition over Bluetooth™ channels.
Ke PengHongbin CaiYaxin ZhangPublished in: Mobility Conference (2007)
Keyphrases
- speech recognition
- histogram equalization
- speech recognition systems
- speaker independent
- speech recognizers
- image enhancement
- contrast enhancement
- hidden markov models
- speech recognizer
- automatic speech recognition
- speech synthesis
- cepstral coefficients
- speech signal
- transfer function
- gray scale
- speech recognition technology
- noisy environments
- acoustic models
- language model
- noisy speech
- pattern recognition
- speaker identification
- speaker adaptation
- speech retrieval
- linear prediction
- color histogram
- gray level
- image contrast
- acoustic features
- image analysis
- machine learning