A Two-Phase Approach for Abstractive Podcast Summarization

TL;DR

This paper introduces a two-phase abstractive summarization method for podcasts, involving sentence selection based on similarity and topic relevance, followed by seq2seq generation, to handle long, colloquial transcripts effectively.

Contribution

It presents a novel two-phase framework specifically designed for podcast summarization, addressing challenges of length and colloquial language with sentence selection and pre-trained models.

Findings

Achieves promising ROUGE scores.

Receives positive human evaluation.

Effectively reduces redundancy and preserves semantics.

Abstract

Podcast summarization is different from summarization of other data formats, such as news, patents, and scientific papers in that podcasts are often longer, conversational, colloquial, and full of sponsorship and advertising information, which imposes great challenges for existing models. In this paper, we focus on abstractive podcast summarization and propose a two-phase approach: sentence selection and seq2seq learning. Specifically, we first select important sentences from the noisy long podcast transcripts. The selection is based on sentence similarity to the reference to reduce the redundancy and the associated latent topics to preserve semantics. Then the selected sentences are fed into a pre-trained encoder-decoder framework for the summary generation. Our approach achieves promising results regarding both ROUGE-based measures and human evaluations.