Unsupervised Topic Segmentation of Meetings with BERT Embeddings

TL;DR

This paper presents an unsupervised method for meeting topic segmentation using BERT embeddings, significantly improving accuracy over previous approaches without requiring annotated data.

Contribution

It introduces a novel unsupervised approach leveraging pre-trained BERT embeddings for meeting topic segmentation, outperforming existing methods.

Findings

15.5% reduction in error rate over previous unsupervised methods

Effective on multiple meeting transcript datasets

Demonstrates the benefit of pre-trained neural architectures in unsupervised tasks

Abstract

Topic segmentation of meetings is the task of dividing multi-person meeting transcripts into topic blocks. Supervised approaches to the problem have proven intractable due to the difficulties in collecting and accurately annotating large datasets. In this paper we show how previous unsupervised topic segmentation methods can be improved using pre-trained neural architectures. We introduce an unsupervised approach based on BERT embeddings that achieves a 15.5% reduction in error rate over existing unsupervised approaches applied to two popular datasets for meeting transcripts.