Robust Front-End Based on MVA and HEQ Post-processing for Arabic Speech Recognition Using Hidden Markov Model Toolkit (HTK).
Elhem TechiniZied SakkaMedsalim BouhlelPublished in: AICCSA (2017)
Keyphrases
- post processing
- hidden markov models
- automatic speech recognition
- speech recognition
- speech corpus
- noisy environments
- speech signal
- phoneme recognition
- preprocessing
- recognition engine
- multi stream
- gesture recognition
- broadcast news
- noisy speech
- hidden state
- hierarchical hidden markov model
- viterbi algorithm
- hidden semi markov models
- handwritten word recognition
- handwriting recognition
- post processed
- hidden states
- filtering method
- markov model
- human experts
- conditional random fields
- computer vision
- continuous hidden markov models
- handwritten words
- speech synthesis
- discriminative training
- sequential data
- image processing