On the advantage of frequency-filtering features for speech recognition with variable sampling frequencies. experiments with speechdatcar databases.
Hermann BauereckerCliment NadeuJaume PadrellPublished in: INTERSPEECH (2003)
Keyphrases
- speech recognition
- cepstral coefficients
- speech recognition systems
- hidden markov models
- pattern recognition
- language model
- noisy environments
- speech understanding
- speech synthesis
- speech processing
- speaker identification
- feature vectors
- automatic speech recognition
- speech signal
- speech recognizer
- feature set
- feature space
- speech recognition technology
- speech retrieval
- data mining
- low level
- feature selection
- computer vision
- classification accuracy
- keyword spotting
- bayesian networks
- feature extraction
- speech recognition errors