AI-Generated Music Detection in Broadcast Monitoring

2026-02-06

AI Summary

This study addresses the significant performance degradation of existing AI-generated music detection methods in broadcast audio, primarily due to real-world challenges such as short music segments and severe speech masking. To bridge this gap, the authors introduce AI-OpenBMAT, the first benchmark dataset specifically designed for AI-generated music detection in broadcast scenarios. They systematically evaluate the robustness of representative models—including CNNs and SpectTTTra—across varying signal-to-noise ratios and segment durations. Experimental results reveal that current approaches achieve F1 scores below 60% under broadcast conditions, highlighting their limited applicability in real-world industrial settings. The proposed dataset thus serves as a critical resource for advancing research and development in this emerging domain.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Generation
• Humans and AI: AI for Accessibility

Application Category

• Web Mining and Content Analysis: Web data quality in the era of algorithmically-generated content
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

AI music generators have advanced to the point where their outputs are often indistinguishable from human compositions. While detection methods have emerged, they are typically designed and validated in music streaming contexts with clean, full-length tracks. Broadcast audio, however, poses a different challenge: music appears as short excerpts, often masked by dominant speech, conditions under which existing detectors fail. In this work, we introduce AI-OpenBMAT, the first dataset tailored to broadcast-style AI-music detection. It contains 3,294 one-minute audio excerpts (54.9 hours) that follow the duration patterns and loudness relations of real television audio, combining human-made production music with stylistically matched continuations generated with Suno v3.5. We benchmark a CNN baseline and state-of-the-art SpectTTTra models to assess SNR and duration robustness, and evaluate on a full broadcast scenario. Across all settings, models that excel in streaming scenarios suffer substantial degradation, with F1-scores dropping below 60% when music is in the background or has a short duration. These results highlight speech masking and short music length as critical open challenges for AI music detection, and position AI-OpenBMAT as a benchmark for developing detectors capable of meeting industrial broadcast requirements.

Problem

• AI-generated music detection
• broadcast audio
• speech masking
• short-duration music
• audio forensics

Innovation

• AI-generated music detection
• broadcast audio
• AI-OpenBMAT dataset
• speech masking
• short-duration music