Combining X-Vectors and Bayesian Batch Active Learning: Two-Stage Active Learning Pipeline for Speech Recognition.
Ognjen KundacinaVladimir VincanDragisa MiskovicPublished in: CoRR (2024)
Keyphrases
- active learning
- speech recognition
- batch mode
- hidden markov models
- automatic speech recognition
- speech recognizer
- speech understanding
- language model
- speech synthesis
- pattern recognition
- learning strategies
- noisy environments
- speech signal
- semi supervised
- learning process
- speech processing
- speaker identification
- speech recognition systems
- speaker dependent
- learning algorithm
- speech recognition technology
- speech recognizers
- feature vectors
- speaker independent
- training set
- bayesian networks
- machine learning
- audio visual speech recognition
- speaker diarization
- speaker recognition
- speech retrieval
- isolated word