Weighting Time-Frequency Representation of Speech Using Auditory Saliency for Automatic Speech Recognition.
Cong-Thanh DoYannis StylianouPublished in: INTERSPEECH (2018)
Keyphrases
- automatic speech recognition
- speech recognition
- speech signal
- word error rate
- broadcast news
- signal processing
- speech corpus
- hidden markov models
- conversational speech
- speech recognizers
- recognition errors
- speech sounds
- spontaneous speech
- speech retrieval
- acoustic features
- language model
- speech synthesis
- phoneme recognition
- computer vision
- speech recognizer
- word recognition
- visual information
- pattern recognition
- neural network
- focus of attention
- tf idf
- error rate
- wavelet transform
- probabilistic model