A new hybrid framework based on Hidden Markov models and K-nearest neighbors for speech recognition.
Samira HazmouneFateh BougamouzaSmaine MazouziMohamed BenmohammedPublished in: Int. J. Speech Technol. (2018)
Keyphrases
- speech recognition
- hidden markov models
- k nearest neighbor
- knn
- automatic speech recognition
- speech signal
- speaker independent
- nearest neighbor
- speech recognizer
- speech synthesis
- speech processing
- continuous hidden markov models
- conditional random fields
- pattern recognition
- keyword spotting
- language model
- speaker identification
- support vector machine
- speaker dependent
- machine learning
- noisy environments
- handwriting recognition
- speech recognition systems
- discriminative training
- data points
- probabilistic model
- visual speech recognition