Using the modulation wavelet transform for feature extraction in automatic speech recognition.
Kenji OkadaTakayuki AraiNoburu KanederuYasunori MomomuraYuji MuraharaPublished in: INTERSPEECH (2000)
Keyphrases
- wavelet transform
- automatic speech recognition
- feature extraction
- speech recognition
- image coding
- hidden markov models
- multiscale
- speech signal
- image compression
- subband
- multiresolution
- broadcast news
- speech retrieval
- word error rate
- wavelet coefficients
- spoken words
- recognition errors
- noisy environments
- conversational speech
- pattern recognition
- frequency domain
- feature vectors
- image processing
- word recognition
- acoustic features
- speech corpus
- feature selection
- computer vision
- spontaneous speech
- speech sounds
- neural network