Knowledge Distillation for Improved Accuracy in Spoken Question Answering
This work addresses accuracy issues in spoken question answering for applications like voice assistants, but it is incremental as it builds on existing knowledge distillation methods.
The paper tackles the problem of noisy automatic speech recognition transcripts limiting spoken question answering accuracy by introducing a knowledge distillation framework that uses language models as supervision to align automatic and manual transcriptions, achieving state-of-the-art results on the Spoken-SQuAD dataset.
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.