Synergy of spectral and perceptual features in multi-source connectionist speech recognition.
Roberto GemelloLoreta MoisaPietro LafacePublished in: INTERSPEECH (2000)
Keyphrases
- multi source
- speech recognition
- data fusion
- information fusion
- cepstral coefficients
- hidden markov models
- speaker adaptation
- speech signal
- neural network
- multiple sources
- expert systems
- speech synthesis
- spectral analysis
- pattern recognition
- automatic speech recognition
- data integration
- speech recognition technology
- language model
- speech recognizer
- data sources
- speech processing
- noisy environments
- speaker independent
- speaker identification
- spectral features
- linear prediction
- artificial intelligence
- speech recognition systems
- isolated word
- hyperspectral images
- community detection
- maximum likelihood
- data mining