Efficient Training of Neural Transducer for Speech Recognition.
Wei ZhouWilfried MichelRalf SchlüterHermann NeyPublished in: INTERSPEECH (2022)
Keyphrases
- speech recognition
- wall street journal corpus
- isolated word
- language model
- automatic speech recognition
- pattern recognition
- hidden markov models
- speech understanding
- acoustic models
- speech synthesis
- speech recognizer
- neural network
- noisy environments
- keyword spotting
- speech recognition technology
- speaker independent
- speech signal
- speech processing
- speaker identification
- speaker adaptation
- speech retrieval
- audio visual speech recognition
- natural language processing
- discriminative training
- speaker recognition
- cepstral coefficients
- speech recognizers
- training set
- feature vectors