Listening Between the Lines: Decoding Podcast Narratives with Language Modeling

2025-11-07

AI Summary

Podcasts’ unscripted, multi-topic, and conversational nature renders existing large language models—pretrained predominantly on news text—ineffective at accurately identifying implicit narrative frames, thereby hindering scalable analysis of public opinion influence. To address this, we propose a fine-grained, human-aligned narrative frame annotation scheme that explicitly anchors frames to concrete semantic entities within dialogue turns and integrates high-level topic modeling to establish systematic linkages among narrative frames, topics, and temporal trends. Leveraging BERT-based fine-tuning, we design a spoken-dialogue-aware frame identification model, achieving a substantial +18.3% F1-score improvement over baselines on real-world podcast data. Our approach transcends the limitations of structured-text pretraining paradigms, offering a scalable, interpretable analytical framework for studying narrative influence in digital media.

Technology Category

• Natural Language Processing: Language Grounding & Multi-modal NLP
• Machine Learning: Large Multimodal Models (LMMs)
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding

Abstract

Podcasts have become a central arena for shaping public opinion, making them a vital source for understanding contemporary discourse. Their typically unscripted, multi-themed, and conversational style offers a rich but complex form of data. To analyze how podcasts persuade and inform, we must examine their narrative structures -- specifically, the narrative frames they employ. The fluid and conversational nature of podcasts presents a significant challenge for automated analysis. We show that existing large language models, typically trained on more structured text such as news articles, struggle to capture the subtle cues that human listeners rely on to identify narrative frames. As a result, current approaches fall short of accurately analyzing podcast narratives at scale. To solve this, we develop and evaluate a fine-tuned BERT model that explicitly links narrative frames to specific entities mentioned in the conversation, effectively grounding the abstract frame in concrete details. Our approach then uses these granular frame labels and correlates them with high-level topics to reveal broader discourse trends. The primary contributions of this paper are: (i) a novel frame-labeling methodology that more closely aligns with human judgment for messy, conversational data, and (ii) a new analysis that uncovers the systematic relationship between what is being discussed (the topic) and how it is being presented (the frame), offering a more robust framework for studying influence in digital media.

Problem

• Analyzing narrative frames in unscripted podcast conversations
• Addressing limitations of existing language models on conversational data
• Developing fine-tuned BERT model to link frames with entities

Innovation

• Fine-tuned BERT model links narrative frames to entities
• Methodology aligns frame-labeling with human judgment
• Analysis reveals relationship between topics and frames