Joint constrained maximum likelihood regression for overlapping speech recognition.
Ken'ichi KumataniRita SinghFriedrich FaubelJohn W. McDonoughYoussef OualilPublished in: ICASSP (2013)
Keyphrases
- speech recognition
- maximum likelihood
- speaker adaptation
- hidden markov models
- speech synthesis
- automatic speech recognition
- speech processing
- language model
- pattern recognition
- noisy environments
- expectation maximization
- speaker identification
- em algorithm
- support vector
- speaker independent
- gaussian mixture model
- speech understanding
- speech recognizer
- additive noise
- speech signal
- gaussian distribution
- speech recognition technology
- speech recognition systems
- speech recognition errors
- discriminative training
- image processing
- keyword spotting
- mixture model
- speech recognizers