Iterative Shallow Fusion of Backward Language Model for End-To-End Speech Recognition.
Atsunori OgawaTakafumi MoriyaNaoyuki KamoNaohiro TawaraMarc DelcroixPublished in: ICASSP (2023)
Keyphrases
- end to end
- speech recognition
- language model
- language modeling
- probabilistic model
- n gram
- information retrieval
- document retrieval
- speech signal
- automatic speech recognition
- test collection
- query expansion
- question answering
- retrieval model
- handwriting recognition
- noisy environments
- query terms
- mixture model
- fusion method
- neural network
- word error rate
- speech recognition systems
- statistical machine translation
- multi modal
- machine learning