Transforming Podcast Preview Generation: From Expert Models to LLM-Based Systems

TL;DR

This paper introduces an LLM-based system for generating podcast previews, demonstrating superior performance and efficiency over traditional expert models through extensive offline and online evaluations.

Contribution

It presents a novel large language model approach for podcast preview generation, reducing reliance on feature engineering and enabling large-scale deployment.

Findings

LLM-generated previews outperform expert model baselines in clarity and interest.

Online A/B testing shows a 4.6% increase in user engagement.

Processing efficiency is increased fivefold compared to traditional methods.

Abstract

Discovering and evaluating long-form talk content such as videos and podcasts poses a significant challenge for users, as it requires a considerable time investment. Previews offer a practical solution by providing concise snippets that showcase key moments of the content, enabling users to make more informed and confident choices. We propose an LLM-based approach for generating podcast episode previews and deploy the solution at scale, serving hundreds of thousands of podcast previews in a real-world application. Comprehensive offline evaluations and online A/B testing demonstrate that LLM-generated previews consistently outperform a strong baseline built on top of various ML expert models, showcasing a significant reduction in the need for meticulous feature engineering. The offline results indicate notable enhancements in understandability, contextual clarity, and interest level, and the online A/B test shows a 4.6% increase in user engagement with preview content, along with a 5x boost in processing efficiency, offering a more streamlined and performant solution compared to the strong baseline of feature-engineered expert models.