Two-dimensional frame-and-feature weighted Viterbi decoding for robust speech recognition.
Yang ChangLin-Shan LeePublished in: ICASSP (2012)
Keyphrases
- speech recognition
- hidden markov models
- noisy environments
- speech synthesis
- speech processing
- viterbi algorithm
- language model
- automatic speech recognition
- speech signal
- cepstral coefficients
- speech recognition technology
- speech understanding
- keyword spotting
- speech recognizer
- pattern recognition
- speech recognition systems
- speaker identification
- speaker independent
- speech retrieval
- audio visual speech recognition
- speaker diarization
- feature vectors
- video frames
- speech recognition errors
- feature space
- feature extraction
- speech recognizers
- isolated word
- information retrieval