arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2001.09727cs.CLcs.SDeess.AS
收藏

基于ConvNets的在线语音识别

Scaling Up Online Speech Recognition Using ConvNets

Facebook AI Research(Facebook人工智能研究院)

Vineel Pratap, Qiantong Xu, Jacob Kahn, Gilad Avidov, Tatiana Likhomanenko, Awni Hannun, Vitaliy Liptchinsky, Gabriel Synnaeve, Ronan Collobert

摘要:

我们设计了一个基于时间深度可分离卷积和连接主义时间分类的在线端到端语音识别系统。我们改进了核心TDS架构,以限制未来的上下文,从而减少延迟,同时保持准确性。该系统的吞吐量几乎是经过良好调整的混合ASR基线的三倍,同时还具有更低的延迟和更好的字出错率。对识别器的效率也很重要的是我们高度优化的束搜索解码器。为了展示我们的设计选择的影响,我们分析了吞吐量、延迟、准确性,并讨论了如何根据用户需求调整这些指标。

英文摘要:

We design an online end-to-end speech recognition system based on Time-Depth Separable (TDS) convolutions and Connectionist Temporal Classification (CTC). We improve the core TDS architecture in order to limit the future context and hence reduce latency while maintaining accuracy. The system has almost three times the throughput of a well tuned hybrid ASR baseline while also having lower latency and a better word error rate. Also important to the efficiency of the recognizer is our highly optimized beam search decoder. To show the impact of our design choices, we analyze throughput, latency, accuracy, and discuss how these metrics can be tuned based on the user requirements.

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。

↑