Sentence Punctuation for Collaborative Commentary Generation in Esports Live-Streaming

TL;DR

This paper introduces two strategies for sentence punctuation in Esports live-streaming commentary, improving the quality of generated collaborative commentary through experiments with a pre-trained language model.

Contribution

The paper proposes novel punctuation strategies for game commentary text sequences and demonstrates their effectiveness over baseline methods.

Findings

Two-sequence punctuation strategy outperforms baseline in automatic metrics.

Fine-tuning a pre-trained language model enhances commentary generation.

Objective and subjective evaluations confirm the strategy's effectiveness.

Abstract

To solve the existing sentence punctuation problem for collaborative commentary generation in Esports live-streaming, this paper presents two strategies for sentence punctuation for text sequences of game commentary, that is, punctuating sentences by two or three text sequence(s) originally punctuated by Youtube to obtain a complete sentence of commentary. We conducted comparative experiments utilizing and fine-tuning a state-of-the-art pre-trained generative language model among two strategies and the baseline to generate collaborative commentary. Both objective evaluations by automatic metrics and subjective analyses showed that our strategy of punctuating sentences by two text sequences outperformed the baseline.