AudioBoost: Increasing Audiobook Retrievability in Spotify Search with Synthetic Query Generation

2025-09-08

AI Summary

Following Spotify’s introduction of audiobooks, a cold-start retrieval bias emerged due to sparse user interactions, rendering conventional search ineffective for exploratory discovery along dimensions such as topic, genre, and era. To address this, we propose an LLM-driven synthetic query generation framework—the first to perform conditional query generation grounded in audiobook metadata—and integrate it into both query auto-completion and retrieval ranking pipelines. Our method jointly optimizes semantic richness and retrieval compatibility, substantially improving the discoverability of newly added content. Offline evaluation demonstrates significant gains in retrieval effectiveness using synthetic queries. Online A/B testing shows a 0.7% increase in audiobook impression rate, a 1.22% lift in click-through rate, and a 1.82% rise in completed exploratory queries. This work delivers a scalable, generative solution for cold-start retrieval of novel content types in streaming platforms.

Technology Category

• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Search and Optimization: Learning to Search
• Machine Learning: Learning Preferences or Rankings

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Topic discovery and tracking

Abstract

Spotify has recently introduced audiobooks as part of its catalog, complementing its music and podcast offering. Search is often the first entry point for users to access new items, and an important goal for Spotify is to support users in the exploration of the audiobook catalog. More specifically, we would like to enable users without a specific item in mind to broadly search by topic, genre, story tropes, decade, and discover audiobooks, authors and publishers they may like. To do this, we need to 1) inspire users to type more exploratory queries for audiobooks and 2) augment our retrieval systems to better deal with exploratory audiobook queries. This is challenging in a cold-start scenario, where we have a retrievabiliy bias due to the little amount of user interactions with audiobooks compared to previously available items such as music and podcast content. To address this, we propose AudioBoost, a system to boost audiobook retrievability in Spotify's Search via synthetic query generation. AudioBoost leverages Large Language Models (LLMs) to generate synthetic queries conditioned on audiobook metadata. The synthetic queries are indexed both in the Query AutoComplete (QAC) and in the Search Retrieval engine to improve query formulation and retrieval at the same time. We show through offline evaluation that synthetic queries increase retrievability and are of high quality. Moreover, results from an online A/B test show that AudioBoost leads to a +0.7% in audiobook impressions, +1.22% in audiobook clicks, and +1.82% in audiobook exploratory query completions.

Problem

• Enhancing audiobook discoverability in Spotify search
• Addressing cold-start bias with synthetic query generation
• Improving retrieval for exploratory audiobook queries

Innovation

• LLM-generated synthetic queries from audiobook metadata
• Indexing synthetic queries in QAC and retrieval systems
• Improving retrievability for cold-start audiobook content