Bypass Temporal Classification: Weakly Supervised Automatic Speech Recognition with Imperfect Transcripts.
Dongji GaoMatthew WiesnerHainan XuLeibny Paola GarciaDaniel PoveySanjeev KhudanpurPublished in: CoRR (2023)
Keyphrases
- automatic speech recognition
- weakly supervised
- speech recognition
- speech signal
- broadcast news
- object class
- pattern recognition
- machine learning
- hidden markov models
- supervised learning
- feature extraction
- support vector
- feature vectors
- image classification
- feature selection
- text classification
- training set
- maximum likelihood
- image features
- object classes
- support vector machine
- viewpoint
- image retrieval
- image processing