Maximizing global entropy reduction for active learning in speech recognition.
Balakrishnan VaradarajanDong YuLi DengAlex AceroPublished in: ICASSP (2009)
Keyphrases
- speech recognition
- active learning
- language model
- hidden markov models
- speech recognition systems
- speech synthesis
- speech recognition technology
- automatic speech recognition
- speech recognizer
- pattern recognition
- speech signal
- speech processing
- speech understanding
- noisy environments
- semi supervised
- learning strategies
- learning algorithm
- speaker recognition
- probabilistic model
- keyword spotting
- training set
- speaker dependent
- speech retrieval
- neural network
- audio visual speech recognition
- speaker identification
- n gram
- signal processing