Dual-channel VTS feature compensation for noise-robust speech recognition on mobile devices.
Iván López-EspejoAntonio M. PeinadoAngel M. GomezJosé A. GonzálezPublished in: IET Signal Process. (2017)
Keyphrases
- speech recognition
- noisy environments
- mobile devices
- dual channel
- speech processing
- speech recognizer
- automatic speech recognition
- digit recognition
- background noise
- hidden markov models
- language model
- speech signal
- speech recognition technology
- speech understanding
- speech synthesis
- speaker identification
- speaker independent
- speech recognition systems
- noisy speech
- noise reduction
- pattern recognition
- speech enhancement
- keyword spotting
- machine learning
- speaker dependent
- speech recognition errors
- cepstral coefficients
- signal to noise ratio
- expectation maximization
- noise level
- mobile learning
- speech recognizers
- isolated word
- neural network
- audio visual speech recognition