ODSQA: Open-domain Spoken Question Answering Dataset
This addresses the understudied issue of spoken question answering for researchers, but is incremental as it focuses on dataset creation and basic improvements.
The paper tackles the problem of machine comprehension of spoken content by releasing the Open-Domain Spoken Question Answering Dataset (ODSQA), the largest real SQA dataset with over three thousand questions, and finds that ASR errors have a catastrophic impact on SQA, with subword units and data augmentation improving performance.
Reading comprehension by machine has been widely studied, but machine comprehension of spoken content is still a less investigated problem. In this paper, we release Open-Domain Spoken Question Answering Dataset (ODSQA) with more than three thousand questions. To the best of our knowledge, this is the largest real SQA dataset. On this dataset, we found that ASR errors have catastrophic impact on SQA. To mitigate the effect of ASR errors, subword units are involved, which brings consistent improvements over all the models. We further found that data augmentation on text-based QA training examples can improve SQA.