Benchmarking LF-MMI, CTC and RNN-T Criteria for Streaming ASR.
Xiaohui ZhangFrank ZhangChunxi LiuKjell SchubertJulian ChanPradyot PrakashJun LiuChing-Feng YehFuchun PengYatharth SarafGeoffrey ZweigPublished in: CoRR (2020)
Keyphrases
- automatic speech recognition
- recurrent neural networks
- nearest neighbor
- speech recognition
- data streams
- speech signal
- real time
- selection criteria
- multiple criteria
- streaming data
- evaluation criteria
- hidden markov models
- class distribution
- speech retrieval
- multi criteria
- noisy environments
- neural network
- real time streaming
- maximization of mutual information
- data streaming
- streaming video
- scalable video
- machine learning
- stream processing
- class labels
- light field
- learning algorithm
- multimedia
- training data
- language model
- logic programs
- test set