Arabic discourse analysis based on acoustic, prosodic and phonetic modeling: elocution evaluation, speech classification and pathological speech correction.
Mohsen MaraouiNaim TerbehMounir ZriguiPublished in: Int. J. Speech Technol. (2018)
Keyphrases
- speech recognition
- speech sounds
- speaker independent
- speech recognition systems
- prosodic features
- speech signal
- speech synthesis
- acoustic models
- text to speech
- automatic speech recognition
- emotional speech
- speech recognizer
- text to speech synthesis
- discourse analysis
- hidden markov models
- feature vectors
- audio visual
- speech segments
- vocal tract
- cooperative
- speaker identification
- emotion recognition
- language model
- spoken document retrieval
- spoken language
- noisy environments
- text classification
- machine learning
- spoken term detection
- computer mediated
- content analysis
- learning strategies