Voice pathology assessment based on automatic speech recognition using Amazigh digits.
Ouissam ZealoukHassan SatoriMohamed HamidiKhalid SatoriPublished in: ICSDE (2018)
Keyphrases
- text to speech
- speech synthesis
- speech recognition
- speech quality
- emotion recognition
- recognition rate
- voice activity detection
- quality assessment
- assessment process
- multi stream
- speech signal
- speech recognition errors
- fully automatic
- synthesized speech
- multi modal
- neural network
- speech music discrimination
- broadcast news
- audio visual
- semi automatic
- data driven
- multimedia