Towards fusion of feature extraction and acoustic model training: a top down process for robust speech recognition.
Yu-Hsiang Bosco ChiuBhiksha RajRichard M. SternPublished in: INTERSPEECH (2009)
Keyphrases
- speech recognition
- feature extraction
- noisy environments
- language model
- speech signal
- speaker identification
- pattern recognition
- speech understanding
- speech processing
- automatic speech recognition
- wall street journal corpus
- hidden markov models
- training process
- speech synthesis
- speech recognizer
- cepstral coefficients
- speech recognition technology
- information retrieval
- wavelet transform
- training data
- image processing
- computer vision
- speech recognition systems
- speaker independent