Pruned RNN-T for fast, memory-efficient ASR training.

Fangjun Kuang Liyong Guo Wei Kang Long Lin Mingshuang Luo Zengwei Yao Daniel Povey

Published in: CoRR (2022)

Keyphrases

memory efficient
recurrent neural networks
external memory
nearest neighbor
training set
test set
automatic speech recognition
multiple sequence alignment
multi dimensional
training samples
training process
iterative deepening
speech recognition