arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2005.08213cs.CLcs.SDeess.AS
收藏

语音文本自适应:一种有效的跨模态提取方法

Speech to Text Adaptation: Towards an Efficient Cross-Modal Distillation

Seoul National University(首尔国立大学);Search Solution Inc.(搜索解决方案公司)

Won Ik Cho, Donghyun Kwak, Ji Won Yoon, Nam Soo Kim

摘要:

言语是最有效的交际手段之一,它所包含的信息有助于表达者思想的传递。然而,主要由于声学特征的繁琐处理,音素或词的后验概率经常被丢弃在理解自然语言。因此,一些最近的口语语言理解(SLU)模块已经利用了保存不确定性信息的端到端结构。这进一步减少了语音识别误差的传播并保证了计算效率。我们认为,在这个过程中,语音理解可以受益于大量的预训练语言模型(LM)的推理。我们将知识从一个具体的基于transformer的文本LM转移到一个SLU模块,该模块可能面临数据短缺,基于最近的跨模态蒸馏方法。我们证明了我们的建议的有效性上的表现流利的语音命令,英语SLU基准。因此,我们通过实验验证了我们的假设,即知识可以从LM的顶层共享到一个完全基于语音的模块,其中抽象的语音预计将满足语义表示。

英文摘要:

Speech is one of the most effective means of communication and is full of information that helps the transmission of utterer's thoughts. However, mainly due to the cumbersome processing of acoustic features, phoneme or word posterior probability has frequently been discarded in understanding the natural language. Thus, some recent spoken language understanding (SLU) modules have utilized end-to-end structures that preserve the uncertainty information. This further reduces the propagation of speech recognition error and guarantees computational efficiency. We claim that in this process, the speech comprehension can benefit from the inference of massive pre-trained language models (LMs). We transfer the knowledge from a concrete Transformer-based text LM to an SLU module which can face a data shortage, based on recent cross-modal distillation methodologies. We demonstrate the validity of our proposal upon the performance on Fluent Speech Command, an English SLU benchmark. Thereby, we experimentally verify our hypothesis that the knowledge could be shared from the top layer of the LM to a fully speech-based module, in which the abstracted speech is expected to meet the semantic representation.

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。

↑