A5068304436- Speech Recognition and Synthesis
- Music and Audio Processing
- Phonetics and Phonology Research
- Intelligent Tutoring Systems and Adaptive Learning
- Advanced Text Analysis Techniques
- Speech and Audio Processing
Central China Normal University
2024
Institut National de la Recherche Scientifique
1990
The motivation of this study is the poor performance speech recognizers on stop consonants. To overcome weakness, word initial and final consonants are modeled at a subphonemic (microsegmental) level. Each consonant segmented into few relatively stationary microsegments: silence, voice bar, burst, aspiration. Microsegments certain phonemically different stops trained together due to their similar spectral properties. Microsegmental models burst aspiration conditioned adjacent vowel category:...
With the popularity of online one-to-one tutoring, there are emerging concerns about quality and effectiveness this kind tutoring. Although some evaluation methods available, they heavily relied on manual coding by experts, which is too costly. Therefore, using machine learning to predict instruction automatically an effective way reduce human costs. Three classification analyzed in paper: 1) Random forest algorithm with engineered descriptive features; 2) LSTM acoustic features generated...