Multi-talker speech recognition under ego-motion noise using Missing Feature Theory.
Gökhan InceKazuhiro NakadaiTobias RodemannHiroshi TsujinoJun-ichi ImuraPublished in: IROS (2010)
Keyphrases
- speech recognition
- noisy environments
- ego motion
- missing data
- hidden markov models
- language model
- noisy speech
- speech signal
- pattern recognition
- automatic speech recognition
- speech recognizer
- field of view
- cepstral coefficients
- speaker identification
- speech synthesis
- speech recognition systems
- speech enhancement
- speaker independent
- noise reduction
- image features
- signal to noise ratio
- background noise
- d scene
- camera motion
- optical flow
- feature vectors
- optic flow
- velocity field
- machine learning
- multiscale
- feature selection
- information retrieval