arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2101.04792eess.AScs.AIcs.LG
收藏

三重丢失下关键词识别的高效学习表示

Learning Efficient Representations for Keyword Spotting with Triplet Loss

Higher IT School, Tomsk State University(托木斯克国立大学高等IT学院);NTR Labs(NTR实验室)

Roman Vygon, Nikolay Mikhaylovskiy

摘要:

在过去的几年里,基于三重丢失的度量嵌入已经成为几个重要的计算机视觉问题的事实上的标准,最值得注意的是,人的重新识别。另一方面,在语音识别领域中,由三元组损失产生的度量嵌入很少用于分类问题。我们填补了这一空白,表明两种表示学习技术的组合:基于三重丢失的嵌入和用于分类的kNN变体,而不是交叉熵损失显着(26%至38%)提高了卷积网络在LibriSpeech衍生的LibriWords数据集上的分类准确性。为此,我们提出了一种新的基于语音相似性的三元组挖掘方法。我们还将目前最好的SOTA for Google Speech Commands数据集V1 10+2类分类提高了约34%,达到98.55%的准确率,V2 10+2类分类提高了约20%,达到98.37%的准确率,V2 35类分类提高了50%以上,达到97.0%的准确率。

英文摘要:

In the past few years, triplet loss-based metric embeddings have become a de-facto standard for several important computer vision problems, most no-tably, person reidentification. On the other hand, in the area of speech recognition the metric embeddings generated by the triplet loss are rarely used even for classification problems. We fill this gap showing that a combination of two representation learning techniques: a triplet loss-based embedding and a variant of kNN for classification instead of cross-entropy loss significantly (by 26% to 38%) improves the classification accuracy for convolutional networks on a LibriSpeech-derived LibriWords datasets. To do so, we propose a novel phonetic similarity based triplet mining approach. We also improve the current best published SOTA for Google Speech Commands dataset V1 10+2 -class classification by about 34%, achieving 98.55% accuracy, V2 10+2-class classification by about 20%, achieving 98.37% accuracy, and V2 35-class classification by over 50%, achieving 97.0% accuracy.

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。

↑