Effective pseudo-relevance feedback for language modeling in extractive speech summarization

Abstract

Extractive speech summarization, aiming to automatically select an indicative set of sentences from a spoken document so as to concisely represent the most important aspects of the document, has become an active area for research and experimentation. An emerging stream of work is to employ the language modeling framework along with the Kullback-Leibler divergence measure for extractive speech summarization, which can perform important sentence selection in an unsupervised manner and has shown preliminary success. This paper presents a continuation of such a general line of research and its main contribution is two-fold. First, by virtue of pseudo-relevance feedback, we explore several effective sentence modeling formulations to enhance the sentence models involved in the LM-based summarization framework. Second, the utilities of our summarization methods and several widely-used methods are analyzed and compared extensively, which demonstrates the effectiveness of our methods.

A margin-based discriminative modeling approach for extractive speech summarization

“…In addition, the difference value is defined as the difference between the minimum and maximum values of the spoken sentence. Table I gives an outline of the different types of features used in this paper, where VSM (Vector Space Model) [21], DLM (Document Likelihood Measure) [22], RM (Relevance Model) [23,24] and SMM (simple mixture model) [24,25] are the relevance values output by the corresponding common unsupervised summarizers, and each is counted as a single summarization (relevance) feature respectively. [26].…”