Efficient Beam Width Control to Suppress Excessive Speech Recognition Computation Time Based on Prior Score Range Normalization.
Satoshi KobashikawaTakaaki HoriYoshikazu YamaguchiTaichi AsamiHirokazu MasatakiSatoshi TakahashiPublished in: INTERSPEECH (2012)
Keyphrases
- speech recognition
- language model
- automatic speech recognition
- hidden markov models
- pattern recognition
- speech recognizer
- speech synthesis
- speech signal
- speech processing
- speaker identification
- speech understanding
- speaker recognition
- noisy environments
- speech recognition technology
- speech recognition errors
- speech retrieval
- handwriting recognition
- non stationary
- natural language processing
- digital video library