VTS feature compensation based on two-layer GMM structure for robust speech recognition.
Lin ZhouHaijing LiYing ChenZhenyang WuYong LuPublished in: WCSP (2016)
Keyphrases
- speech recognition
- speaker identification
- noisy environments
- speech signal
- speaker recognition
- hidden markov models
- language model
- automatic speech recognition
- speech processing
- cepstral coefficients
- speech synthesis
- speech recognition technology
- speech understanding
- gaussian mixture model
- voice activity detection
- pattern recognition
- speech recognizer
- speech recognition systems
- speaker independent
- speech retrieval
- speech recognizers
- isolated word
- machine learning
- feature vectors
- mel frequency cepstral coefficients
- speaker adaptation
- computer vision
- speech recognition errors
- information retrieval