Exploring Joint Equalization of Spatial-Temporal Contextual Statistics of Speech Features for Robust Speech Recognition.
Hsin-Ju HsiehJeih-Weih HungBerlin ChenPublished in: INTERSPEECH (2012)
Keyphrases
- speech recognition
- spatial temporal
- speech recognition systems
- speech signal
- noisy environments
- speech synthesis
- automatic speech recognition
- hidden markov models
- cepstral coefficients
- language model
- speech processing
- speech recognizer
- speaker identification
- temporal information
- isolated word
- speaker independent
- speech recognition technology
- speech recognizers
- pattern recognition
- speaker recognition
- action recognition
- mel frequency cepstral coefficients
- spatial and temporal
- spatio temporal
- feature space
- speech recognition errors
- low level
- feature vectors
- contextual information
- feature extraction
- speaker diarization
- image features
- speaker dependent
- machine learning