Hybrid Input-type Recurrent Neural Network Language Modeling for End-to-end Speech Recognition.
Phuttapong SertsiPoonlap LamsrichanVataya ChunwijitraManabu OkumuraPublished in: JCSSE (2021)
Keyphrases
- end to end
- speech recognition
- language model
- language modeling
- recurrent neural networks
- probabilistic model
- n gram
- query expansion
- automatic speech recognition
- document retrieval
- feed forward
- retrieval model
- speech signal
- information retrieval
- cross lingual
- neural network
- pattern recognition
- word error rate
- test collection
- mixture model
- hidden markov models
- handwriting recognition
- speech recognition systems
- query terms
- image processing
- computer vision
- search engine
- learning algorithm