Spoken Language Identification in Unseen Target Domain Using Centroid Similarity Loss With Adaptive Gradient Blending.
Muralikrishna HSujeet KumarDileep Aroor DineshVeena ThenkanidiyoorPublished in: IEEE Access (2024)
Keyphrases
- language identification
- domain adaptation
- target domain
- transfer learning
- cross domain
- labeled data
- speaker identification
- similarity measure
- speech recognition
- training data
- document images
- domain specific
- machine learning
- semantic similarity
- multiple sources
- broadcast news
- data points
- edge detection
- hidden markov models
- automatic speech recognition