Multi-Source Domain Adaptation for Text-Independent Forensic Speaker Recognition.
Zhenyu WangJohn H. L. HansenPublished in: IEEE ACM Trans. Audio Speech Lang. Process. (2022)
Keyphrases
- domain adaptation
- multi source
- multiple sources
- speaker recognition
- gaussian mixture model
- vector quantization
- data fusion
- data sources
- target domain
- information fusion
- speaker identification
- information retrieval
- transfer learning
- speech signal
- cross domain
- web documents
- database
- feature extraction
- sentiment classification
- data mining
- databases
- text documents
- text data
- face recognition
- semi supervised learning
- labeled data
- text mining
- bayesian networks