Automatic Summarization of Open-Domain Multiparty Dialogues in Diverse Genres

Abstract

Automatic summarization of open-domain spoken dialogues is a relatively new research area. This article introduces the task and the challenges involved and motivates and presents an approach for obtaining automatic-extract summaries for human transcripts of multiparty dialogues of four different genres, without any restriction on domain.We address the following issues, which are intrinsic to spoken-dialogue summarization and typically can be ignored when summarizing written text such as news wire data: (1) detection and removal of speech disfluencies; (2) detection and insertion of sentence boundaries; and (3) detection and linking of cross-speaker information units (question-answer pairs).A system evaluation is performed using a corpus of 23 dialogue excerpts with an average duration of about 10 minutes, comprising 80 topical segments and about 47,000 words total. The corpus was manually annotated for relevant text spans by six human annotators. The global evaluation shows that for the two more informal genres, our summarization system using dialoguespecific components significantly outperforms two baselines: (1) a maximum-marginal-relevance ranking algorithm using TF*IDF term weighting, and (2) a LEAD baseline that extracts the first n words from a text.

Automatic analysis of multiparty meetings

“…A fragment may be a speaker turn, a dialogue act or a "speech spurt" (a sequence of speech from a single speaker bounded by silence, but not not necessarily corresponding to a linguistic unit). The algorithms that we have used for extractive summarisation rely strongly on speech transcripts, and employ methods such as tf ·idf used in text retrieval (Zechner 2002). However, text-only methods for meeting summarisation can be significantly improved by considering information related to speaker turns and to prosody (Murray et al 2006).…”

Abstractive Summarization of Voice Communications

“…The problem of automatic speech summarization was initially investigated in the 80' in the context of several DARPA projects [3]. Automatic summarization of meetings has been typically approached in a so-called "extractive" fashion that is by extracting excerpts of the dialogs and by assembling them into a hopefully coherent text [4,5].…”

Using Question-Answer Pairs in Extractive Summarization of Email Conversations

“…However, there has been some work on summarizing meetings that bears some relation to ours. (Zechner, 2002), for example, presents a meeting summarization system which uses the MMR algorithm to find sentences that are most salient while minimizing the redundancy in the summary. The similarity weights in the MMR algorithm are modified using three features, including whether a sentence belongs to a question-answer pair.…”