Current ASR systems are mainly trained and evaluated at the utterance level. Long range cross utterance context can be incorporated. A key task is to derive a suitable compact representation of the most relevant history contexts. In contrast to previous researches based on either LSTM-RNN encoded histories that attenuate the information from longer range contexts, or frame level concatenation of transformer context embeddings, in this paper compact low-dimensional cross utterance contextual features are learned in the Conformer-Transducer Encoder using specially designed attention pooling layers that are applied over efficiently cached preceding utterances history vectors. Experiments on the 1000-hr Gigaspeech corpus demonstrate that the proposed contextualized streaming Conformer-Transducers outperform the baseline using utterance internal context only with statistically significant WER reductions of 0.7% to 0.5% absolute (4.3% to 3.1% relative) on the dev and test data.
@article{arxiv.2306.13307,
title = {Towards Effective and Compact Contextual Representation for Conformer Transducer Speech Recognition Systems},
author = {Mingyu Cui and Jiawen Kang and Jiajun Deng and Xi Yin and Yutao Xie and Xie Chen and Xunying Liu},
journal= {arXiv preprint arXiv:2306.13307},
year = {2023}
}