arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2108.02598cs.CLcs.HC
收藏

从BERT Transformer到语音Transformer的意图分类知识蒸馏

Knowledge Distillation from BERT Transformer to Speech Transformer for Intent Classification

National University of Singapore(新加坡国立大学)

Yidi Jiang, Bidisha Sharma, Maulik Madhavi, Haizhou Li

摘要:

使用语音的端到端意图分类与使用自动语音识别(ASR)、随后是自然语言处理模块的常规流水线方法相比具有许多优点。它试图在不使用中间ASR模块的情况下从语音中预测意图。然而,这样的端到端框架遭受在口语语言理解中具有较高声学变化的大语音资源的不可用。在这项工作中,我们利用范围的Transformer蒸馏方法,专门设计用于知识蒸馏从Transformer为基础的语言模型的Transformer为基础的语音模型。在这方面,我们利用可靠的和广泛使用的双向编码器表示从Transformers(BERT)模型作为语言模型和转移的知识,以建立一个声学模型使用语音的意图分类。特别是,一个多级Transformer为基础的教师-学生模型的设计,和知识蒸馏进行跨注意和隐藏的子层的不同Transformer层的学生和教师模型。我们实现了一个意图分类准确率为99.10%和88.79%的流利的语音语料库和ATIS数据库,分别。此外,所提出的方法表现出更好的性能和鲁棒性在声学退化的条件下相比,基线方法。

英文摘要:

End-to-end intent classification using speech has numerous advantages compared to the conventional pipeline approach using automatic speech recognition (ASR), followed by natural language processing modules. It attempts to predict intent from speech without using an intermediate ASR module. However, such end-to-end framework suffers from the unavailability of large speech resources with higher acoustic variation in spoken language understanding. In this work, we exploit the scope of the transformer distillation method that is specifically designed for knowledge distillation from a transformer based language model to a transformer based speech model. In this regard, we leverage the reliable and widely used bidirectional encoder representations from transformers (BERT) model as a language model and transfer the knowledge to build an acoustic model for intent classification using the speech. In particular, a multilevel transformer based teacher-student model is designed, and knowledge distillation is performed across attention and hidden sub-layers of different transformer layers of the student and teacher models. We achieve an intent classification accuracy of 99.10% and 88.79% for Fluent speech corpus and ATIS database, respectively. Further, the proposed method demonstrates better performance and robustness in acoustically degraded condition compared to the baseline method.

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。

↑