arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1707.07413cs.CLcs.NE
收藏

探索用于端到端语音识别的神经传感器

Exploring Neural Transducers for End-to-End Speech Recognition

Baidu Silicon Valley AI Lab(百度硅谷人工智能实验室)

Eric Battenberg, Jitong Chen, Rewon Child, Adam Coates, Yashesh Gaur, Yi Li, Hairong Liu, Sanjeev Satheesh, David Seetapun, Anuroop Sriram, Zhenyao Zhu

摘要:

在这项工作中,我们进行了CTC,RNN传感器,和基于注意力的Seq 2Seq模型之间的经验比较,用于端到端的语音识别。我们表明,在没有任何语言模型的情况下,Seq 2Seq和RNN-Transducer模型在流行的Hub 5 '00基准测试中的表现都优于具有语言模型的最佳CTC模型。在我们内部的多样化数据集上,这些趋势仍在继续-在束搜索之后,使用语言模型重新评分的RNNTransducer模型优于我们最好的CTC模型。这些结果简化了语音识别流水线,使得解码现在可以纯粹地表示为神经网络操作。我们还研究了编码器架构的选择如何影响这三种模型的性能-当所有编码器层都只向前时,以及当编码器对输入表示进行下采样时。

英文摘要:

In this work, we perform an empirical comparison among the CTC, RNN-Transducer, and attention-based Seq2Seq models for end-to-end speech recognition. We show that, without any language model, Seq2Seq and RNN-Transducer models both outperform the best reported CTC models with a language model, on the popular Hub5'00 benchmark. On our internal diverse dataset, these trends continue - RNNTransducer models rescored with a language model after beam search outperform our best CTC models. These results simplify the speech recognition pipeline so that decoding can now be expressed purely as neural network operations. We also study how the choice of encoder architecture affects the performance of the three models - when all encoder layers are forward only, and when encoders downsample the input representation aggressively.

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。

↑