Mixed Precision Quantization of Transformer Language Models for Speech Recognition.
Junhao XuShoukang HuJianwei YuXunying LiuHelen MengPublished in: ICASSP (2021)
Keyphrases
- speech recognition
- language model
- language modeling
- probabilistic model
- n gram
- information retrieval
- document retrieval
- mixture model
- test collection
- query expansion
- speech signal
- speech recognizer
- retrieval model
- automatic speech recognition
- noisy environments
- speech synthesis
- smoothing methods
- word error rate
- average precision
- speaker identification
- speech recognition systems
- translation model
- handwriting recognition
- pattern recognition
- vector space model
- language independent
- neural network
- non stationary
- query terms
- feature selection