Introducing temporal asymmetries in feature extraction for automatic speech recognition.
Garimella S. V. S. SivaramHynek HermanskyPublished in: INTERSPEECH (2008)
Keyphrases
- automatic speech recognition
- feature extraction
- speech recognition
- speech signal
- word error rate
- hidden markov models
- broadcast news
- speech retrieval
- conversational speech
- temporal information
- noisy environments
- feature vectors
- recognition errors
- spontaneous speech
- feature selection
- wavelet transform
- pattern recognition
- spoken words
- speaker identification
- speech corpus
- extracted features
- frequency domain
- image classification
- feature set
- feature space
- bayesian networks
- face recognition
- image processing
- word recognition