Automatic Title Generation for Spoken Broadcast News

摘要

In this paper, we implemented a set of title generation methods using training set of 21,190 news stories and evaluated them on an independent test corpus of 1,006 broadcast news documents, comparing the results over manual transcription to the results over automatically recognized speech. We use both F1 and the average number of correct title words in the correct order as metric. Overall, the results show that title generation for speech recognized news documents is possible at a level approaching the accuracy of titles generated for perfect text transcriptions.