End-to-End Training of Acoustic Models for Large Vocabulary Continuous Speech Recognition with TensorFlow.
Ehsan VarianiTom BagbyErik McDermottMichiel BacchianiPublished in: INTERSPEECH (2017)
Keyphrases
- end to end
- acoustic models
- speech recognizer
- automatic speech recognition
- speech recognition
- spoken language
- broadcast news
- admission control
- hidden markov models
- congestion control
- discriminative training
- neural network
- speech signal
- noisy environments
- language model
- speaker independent
- text localization and recognition
- application layer
- training set
- natural language
- training data