arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2007.13802eess.AScs.CLcs.LGcs.SD
收藏

用于端到端语音识别的RNN转换器的有效最小字错误率训练

Efficient minimum word error rate training of RNN-Transducer for end-to-end speech recognition

Jinxi Guo, Gautam Tiwari, Jasha Droppo, Maarten Van Segbroeck, Che-Wei Huang, Andreas Stolcke, Roland Maas

摘要:

在这项工作中,我们提出了一种新的和有效的最小字错误率(MWER)的RNN传感器(RNN-T)的训练方法。与以前关于这个主题的工作不同,它执行动态有限大小的波束搜索解码并为预期的编辑距离计算生成对齐分数,在我们提出的方法中,我们重新计算并汇总N个最佳列表中每个假设的所有可能对齐的分数。使用前向-后向算法有效地计算假设概率得分和反向传播梯度。此外,所提出的方法允许我们解耦的解码和训练过程,因此我们可以执行离线并行解码和MWER训练的每个子集迭代。实验结果表明,这种半实时方法可以将实时方法的速度提高6倍,并且与基线RNN-T模型相比,WER提高了3.6%。当为端点添加EOS时,所提出的MWER训练还可以有效地减少RNN-T模型引入的高删除错误(9.2%WER减少)。如果我们使用提出的RNN-T重新评分方法对假设进行重新排序,并使用外部RNN-LM执行额外的重新评分,则可以实现进一步的改进。最好的系统实现了5%的英语测试集的真实远场录音和11.6%的WER减少音乐领域的话语相对改善。

英文摘要:

In this work, we propose a novel and efficient minimum word error rate (MWER) training method for RNN-Transducer (RNN-T). Unlike previous work on this topic, which performs on-the-fly limited-size beam-search decoding and generates alignment scores for expected edit-distance computation, in our proposed method, we re-calculate and sum scores of all the possible alignments for each hypothesis in N-best lists. The hypothesis probability scores and back-propagated gradients are calculated efficiently using the forward-backward algorithm. Moreover, the proposed method allows us to decouple the decoding and training processes, and thus we can perform offline parallel-decoding and MWER training for each subset iteratively. Experimental results show that this proposed semi-on-the-fly method can speed up the on-the-fly method by 6 times and result in a similar WER improvement (3.6%) over a baseline RNN-T model. The proposed MWER training can also effectively reduce high-deletion errors (9.2% WER-reduction) introduced by RNN-T models when EOS is added for endpointer. Further improvement can be achieved if we use a proposed RNN-T rescoring method to re-rank hypotheses and use external RNN-LM to perform additional rescoring. The best system achieves a 5% relative improvement on an English test-set of real far-field recordings and a 11.6% WER reduction on music-domain utterances.

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。

↑