LongFNT: Long-form Speech Recognition with Factorized Neural Transducer.
Xun GongYu WuJinyu LiShujie LiuRui ZhaoXie ChenYanmin QianPublished in: CoRR (2022)
Keyphrases
- speech recognition
- hidden markov models
- language model
- speech synthesis
- speech understanding
- speech processing
- pattern recognition
- speech signal
- automatic speech recognition
- handwriting recognition
- speech recognizer
- neural network
- speech recognition systems
- speech recognition technology
- cepstral coefficients
- keyword spotting
- speaker recognition
- noisy environments
- speech recognition errors
- speaker dependent
- speech retrieval
- isolated word
- speaker independent
- speaker identification
- non stationary
- bayesian networks
- feature selection
- information retrieval