An exploration of semi-supervised and language-adversarial transfer learning using hybrid acoustic model for hindi speech recognition.
Ankit KumarRajesh Kumar AggarwalPublished in: J. Reliab. Intell. Environ. (2022)
Keyphrases
- speech recognition
- transfer learning
- semi supervised
- labeled data
- semi supervised learning
- unlabeled data
- named entity recognition
- isolated word
- active learning
- knowledge transfer
- hidden markov models
- speech synthesis
- indian languages
- language model
- cross lingual
- target language
- cross domain
- automatic speech recognition
- speech recognizer
- pairwise
- supervised learning
- pattern recognition
- speech signal
- speaker identification
- labeled and unlabeled data
- natural language
- speech recognition systems
- reinforcement learning
- unsupervised learning
- training data
- transfer knowledge
- language identification
- collaborative filtering
- text classification
- machine learning algorithms
- machine learning
- text categorization
- co training
- text mining
- language processing
- noisy environments
- machine translation
- k nearest neighbor
- data points
- learning algorithm
- target domain
- multi task
- prior knowledge
- artificial intelligence