arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2003.10369eess.AS
收藏

基于Scout网络的低延迟流媒体语音识别

Low Latency End-to-End Streaming Speech Recognition with a Scout Network

Chengyi Wang, Yu Wu, Shujie Liu, Jinyu Li, Liang Lu, Guoli Ye, Ming Zhou

摘要:

基于注意力的Transformer模型已经在离线模式下的语音识别(SR)中取得了很好的效果。然而,在流式传输模式中,当在每个编码器层中应用固定长度的前瞻窗口时,Transformer模型通常招致显著的延迟以维持其识别准确性。针对Transformer模型,提出了一种新的低延迟流方法,该方法由侦察网络和识别网络组成。侦察网络检测整个词的边界,而看不到任何未来的帧,而识别网络预测下一个子词通过利用预测的边界之前的所有帧的信息。我们的模型实现了最好的性能(2.7/6.4 WER),只有639 ms的延迟上的测试清洁和测试其他数据集的Librispeech。

英文摘要:

The attention-based Transformer model has achieved promising results for speech recognition (SR) in the offline mode. However, in the streaming mode, the Transformer model usually incurs significant latency to maintain its recognition accuracy when applying a fixed-length look-ahead window in each encoder layer. In this paper, we propose a novel low-latency streaming approach for Transformer models, which consists of a scout network and a recognition network. The scout network detects the whole word boundary without seeing any future frames, while the recognition network predicts the next subword by utilizing the information from all the frames before the predicted boundary. Our model achieves the best performance (2.7/6.4 WER) with only 639 ms latency on the test-clean and test-other data sets of Librispeech.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑