QASR: Spoken Question Answering Using Semantic Role Labeling

In this demo, we will present QASR- a question answering system that uses semantic role labeling. Given a question, QASR semantically parses it, formulates a query, sends it to a search engine (in this case, Google), and semantically parses the candidate answers matching the semantic argument in question. QASR supports multiple modalities for question answering. It is possible to write down the question using a web or command-prompt interface, or speak the question via speech interface. Question answering (QA) is the task of finding a concise answer to a Natural Language question. Question answering system uses a search engine but differs from a web search task by the type of its input and output. Input to a search engine is a query while a QA system takes a natural language question as an input. For example, if one wants to find out “Who first broke the sound barrier?”, the user of a search engine would convert the question into a query like first AND broke AND “sound barrier”, while the user of a QA system would type or say the question. The output of the QA system is a concise answer: Yeager, in the case of the above question while the search engine produces snippets with link to the web pages. In this work we only focus on factoid questions. Answer to a factoid question is a single word or a short phrase. The above example is a factoid question, while “What is the meaning of life ” is not. Factoid QA has been extensivel