论文标题
知识蒸馏以提高口头问题的准确性
Knowledge Distillation for Improved Accuracy in Spoken Question Answering
论文作者
论文摘要
口头问题回答（SQA）是一项具有挑战性的任务，需要机器充分了解复杂的口语文档。自动语音识别（ASR）在质量检查系统的开发中起着重要作用。但是，最近的工作表明，ASR系统会生成高度嘈杂的成绩单，这严重限制了机器对SQA任务的理解能力。为了解决这个问题，我们提出了一个新颖的蒸馏框架。具体来说，我们制定了一种培训策略来从口头文件和书面同行中执行知识蒸馏（KD）。我们的工作迈出了从语言模型中提取知识的一步，作为监督信号，通过降低自动转录和手动转录之间的不对对准，从而提高学生的准确性。实验表明，我们的方法的表现优于口语数据集上的几种最先进的语言模型。
Spoken question answering (SQA) is a challenging task that requires the machine to fully understand the complex spoken documents. Automatic speech recognition (ASR) plays a significant role in the development of QA systems. However, the recent work shows that ASR systems generate highly noisy transcripts, which critically limit the capability of machine comprehension on the SQA task. To address the issue, we present a novel distillation framework. Specifically, we devise a training strategy to perform knowledge distillation (KD) from spoken documents and written counterparts. Our work makes a step towards distilling knowledge from the language model as a supervision signal to lead to better student accuracy by reducing the misalignment between automatic and manual transcriptions. Experiments demonstrate that our approach outperforms several state-of-the-art language models on the Spoken-SQuAD dataset.
