Speech Vs. Transcript: Does It Matter For Human Annotators In Speech Summarization?

Abstract

Reference summaries for abstractive speech summarization require human annotation, which can be performed by listening to an audio recording or by reading textual transcripts of the recording. In this paper, we examine whether summaries based on annotators listening to the recordings differ from those based on annotators reading transcripts. Using existing intrinsic evaluation based on human evaluation, automatic metrics, LLM-based evaluation, and a retrieval-based reference-free method. We find that summaries are indeed different based on the source modality, and that speech-based summaries are more factually consistent and information-selective than transcript-based summaries. Meanwhile, transcript-based summaries are impacted by recognition errors in the source, and expert-written summaries are more informative and reliable. We make all the collected data and analysis code public(https://github.com/cmu-mlsp/interview_humanssum) to facilitate the reproduction of our work and advance

Stats

• citations2
• S2 citations—
• github stars4
• HF likes0
• heat score4.98
• arxiv keysharma2024speech

Code

• cmu-mlsp/interview_humanssum—

Related papers

• Augsumm: Towards Generalizable Speech Summarization Using Synthetic Labels From Large Language Model (2024)4.53
• Toward Unifying Text Segmentation And Long Document Summarization (2022)8.60
• \'etude De L'informativit\'e Des Transcriptions : Une Approche Bas\'ee Sur Le R\'esum\'e Automatique (2018)0.00
• Unsupervised Broadcast News Summarization; A Comparative Study On Maximal Marginal Relevance (MMR) And Latent Semantic Analysis (LSA) (2023)5.24
• Vt-ssum: A Benchmark Dataset For Video Transcript Segmentation And Summarization (2021)2.76
• Streamhover: Livestream Transcript Summarization And Annotation (2021)7.16
• Team MTS @ Automin 2021: An Overview Of Existing Summarization Approaches And Comparison To Unsupervised Summarization Techniques (2024)0.00
• The Impact Of Automatic Speech Transcription On Speaker Attribution (2025)3.58