Features based on filtering and spectral peaks in autocorrelation domain for robust speech recognition.
Gholamreza FarahaniSeyed Mohammad AhadiMohammad Mehdi HomayounpourPublished in: Comput. Speech Lang. (2007)
Keyphrases
- speech recognition
- cepstral coefficients
- speech recognition systems
- noisy environments
- speech processing
- speech signal
- pattern recognition
- language model
- speech synthesis
- speech understanding
- hidden markov models
- feature set
- automatic speech recognition
- low level
- speech recognition technology
- speaker identification
- non stationary
- feature vectors
- feature extraction
- computer vision
- image processing
- speech recognition errors
- speech retrieval
- classification accuracy
- spectral analysis
- speaker recognition
- feature space
- speech recognizer
- information retrieval
- neural network