arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2005.03191eess.AScs.CLcs.LGcs.SD
收藏

ContextNet:改进卷积神经网络用于全局上下文自动语音识别

ContextNet: Improving Convolutional Neural Networks for Automatic Speech Recognition with Global Context

Google Inc.(谷歌公司)

Wei Han, Zhengdong Zhang, Yu Zhang, Jiahui Yu, Chung-Cheng Chiu, James Qin, Anmol Gulati, Ruoming Pang, Yonghui Wu

摘要:

卷积神经网络(CNN)在端到端语音识别方面已经显示出有希望的结果,尽管在性能上仍然落后于其他最先进的方法。在本文中,我们研究如何弥合这一差距,并超越一种新的CNN-RNN-transducer架构,我们称之为ContextNet。ContextNet具有完全卷积编码器,通过添加挤压和激励模块将全局上下文信息纳入卷积层。此外,我们提出了一个简单的缩放方法,缩放的ContextNet的宽度,实现良好的计算和准确性之间的权衡。我们证明,在广泛使用的LibriSpeech基准测试中,ContextNet在没有外部语言模型(LM)的情况下实现了2.1%/4.6%的单词错误率(WER),LM为1.9%/4.1%,在干净/嘈杂的LibriSpeech测试集上只有10 M参数时为2.9%/7.0%。相比之下,之前发布的最佳系统在LM参数下为2.0%/4.6%,在20 M参数下为3.9%/11.3%。所提出的ContextNet模型的优越性也在一个更大的内部数据集上得到了验证。

英文摘要:

Convolutional neural networks (CNN) have shown promising results for end-to-end speech recognition, albeit still behind other state-of-the-art methods in performance. In this paper, we study how to bridge this gap and go beyond with a novel CNN-RNN-transducer architecture, which we call ContextNet. ContextNet features a fully convolutional encoder that incorporates global context information into convolution layers by adding squeeze-and-excitation modules. In addition, we propose a simple scaling method that scales the widths of ContextNet that achieves good trade-off between computation and accuracy. We demonstrate that on the widely used LibriSpeech benchmark, ContextNet achieves a word error rate (WER) of 2.1%/4.6% without external language model (LM), 1.9%/4.1% with LM and 2.9%/7.0% with only 10M parameters on the clean/noisy LibriSpeech test sets. This compares to the previous best published system of 2.0%/4.6% with LM and 3.9%/11.3% with 20M parameters. The superiority of the proposed ContextNet model is also verified on a much larger internal dataset.

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。

↑