arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2210.00367eess.AScs.CL
收藏

Transformer、卷积和递归神经网络在音素识别中的比较

A Comparison of Transformer, Convolutional, and Recurrent Neural Networks on Phoneme Recognition

Seoul National University(首尔国立大学)

Kyuhong Shim, Wonyong Sung

摘要:

音素识别是语音识别的一个重要组成部分,它要求能够从多帧中提取语音特征。在本文中,我们比较和分析CNN,RNN,Transformer和Conformer模型使用音素识别。对于CNN,使用ContextNet模型进行实验。首先,我们比较了不同的约束条件下,如感受野长度,参数大小和层深度的各种架构的准确性。其次,我们解释了这些模型的性能差异,特别是当可观测序列长度变化时。我们的分析表明,Transformer和Conformer模型受益于通过输入帧的自我注意的长距离可达性。

英文摘要:

Phoneme recognition is a very important part of speech recognition that requires the ability to extract phonetic features from multiple frames. In this paper, we compare and analyze CNN, RNN, Transformer, and Conformer models using phoneme recognition. For CNN, the ContextNet model is used for the experiments. First, we compare the accuracy of various architectures under different constraints, such as the receptive field length, parameter size, and layer depth. Second, we interpret the performance difference of these models, especially when the observable sequence length varies. Our analyses show that Transformer and Conformer models benefit from the long-range accessibility of self-attention through input frames.

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。

↑