Robust Features for Speaker-Independent Speech Recognition Based on a Certain Class of Translation-Invariant Transformations.
Florian MüllerAlfred MertinsPublished in: NOLISP (2009)
Keyphrases
- speech recognition
- speaker independent
- speech recognition systems
- translation invariant
- speaker dependent
- noisy environments
- hidden markov models
- speaker identification
- speech recognizer
- language model
- automatic speech recognition
- neural network
- speech signal
- speech synthesis
- denoising
- pattern recognition
- feature extraction
- mel frequency cepstral coefficients
- co occurrence
- feature vectors
- extracting features
- digit recognition
- feature selection