Leveraging Data Collection and Unsupervised Learning for Code-Switched Tunisian Arabic Automatic Speech Recognition.
Ahmed Amine Ben AbdallahAta KabboudiAmir KanounSalah ZaiemPublished in: ICASSP (2024)
Keyphrases
- automatic speech recognition
- data collection
- unsupervised learning
- speech corpus
- speech recognition
- speech signal
- word error rate
- hidden markov models
- data entry
- broadcast news
- speech retrieval
- data analysis
- supervised learning
- conversational speech
- spoken words
- sensor networks
- handwriting recognition
- noisy environments
- recognition errors
- text classification
- compound words
- semi supervised
- word recognition
- wireless sensor networks
- data mining
- expectation maximization
- acoustic features
- pattern recognition
- machine learning
- feature extraction
- neural network
- object recognition
- computer vision
- spontaneous speech
- active learning
- question answering
- language model