On the interplay between auditory-based features and locally recurrent neural networks for robust speech recognition in noise.
Jürgen TchorzKlaus KasperHerbert ReiningerBirger KollmeierPublished in: EUROSPEECH (1997)
Keyphrases
- speech recognition
- recurrent neural networks
- noisy environments
- speech recognition systems
- speech processing
- automatic speech recognition
- pattern recognition
- hidden markov models
- signal processing
- speaker identification
- feed forward
- speech signal
- cepstral coefficients
- speech synthesis
- language model
- feature vectors
- feature space
- speech recognizer
- neural network
- recurrent networks
- feature set
- mel frequency cepstral coefficients
- feature extraction
- computer vision
- noisy speech
- background noise
- noise level
- noise reduction
- maximum likelihood
- data mining