SemEval 2022 Task 12: Symlink- Linking Mathematical Symbols to their Descriptions

TL;DR

This paper introduces a new annotated corpus for punctuation restoration in livestreaming video transcripts, highlighting the challenges and limitations of existing NLP tools in this domain.

Contribution

It provides the BehancePR corpus and demonstrates the inadequacy of current NLP tools for punctuation restoration in livestreaming transcripts.

Findings

Existing NLP tools struggle with non-punctuated livestreaming transcripts.

The BehancePR corpus reveals the complexity of punctuation restoration in this domain.

Current models are insufficient, indicating a need for more robust solutions.

Abstract

Given the increasing number of livestreaming videos, automatic speech recognition and post-processing for livestreaming video transcripts are crucial for efficient data management as well as knowledge mining. A key step in this process is punctuation restoration which restores fundamental text structures such as phrase and sentence boundaries from the video transcripts. This work presents a new human-annotated corpus, called BehancePR, for punctuation restoration in livestreaming video transcripts. Our experiments on BehancePR demonstrate the challenges of punctuation restoration for this domain. Furthermore, we show that popular natural language processing toolkits are incapable of detecting sentence boundary on non-punctuated transcripts of livestreaming videos, calling for more research effort to develop robust models for this area.