Tigrinya End-to-End Speech Recognition: A Hybrid Connectionist Temporal Classification-Attention Approach.
Bereket Desbele GhebregiorgisYonatan Yosef TekleMebrahtu Fisshaye KidaneMussie Kaleab KeletaRutta Fissehatsion GhebraebDaniel Tesfai GebretatiosPublished in: PanAfriCon AI (1) (2023)
Keyphrases
- end to end
- speech recognition
- pattern recognition
- automatic speech recognition
- language model
- text localization and recognition
- speech recognition technology
- speech synthesis
- speaker identification
- noisy environments
- speech processing
- speech signal
- hidden markov models
- text classification
- congestion control
- speech recognizer
- neural network
- support vector machine
- feature space
- feature extraction
- speaker diarization
- speech recognition systems
- speech recognizers
- feature set
- speaker adaptation
- feature selection
- machine learning