发表机构
Deezer Research（Deezer 研究院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对混合音乐中AI生成音轨的检测难题，提出不依赖音轨分层的框架，利用inspectrogram与维纳滤波器结合CNN模型，在高频源上表现优异，但低频贝斯检测受限，指出源分离是主要瓶颈。
AI 中文摘要
生成音频在音乐制作过程中的引入导致混合音乐曲目增多，这些曲目将真实的人类表演与AI生成的音轨分层（stems）相融合，对在二元设置下运行的传统AI音乐检测器构成了挑战。在这项工作中，我们提出了一种不依赖音轨分层（stem-agnostic）的框架，用于识别混合音乐混合物中的合成音频源。我们引入了inspectrogram，一种新颖的时频表示，它映射了音频频谱上合成内容的局部概率。通过将inspectrogram与估计目标音轨能量主导度的维纳滤波器（Wiener filter）相结合，一个单一的CNN模型评估特定音轨是否由生成模型产生。该模型在渲染的混合混合物上训练，并在各种音轨类别上进行评估，在高频源（如人声、鼓和吉他）上取得了强劲性能，但在低频、窄带的贝斯上表现不佳。我们得出结论，分离质量影响检测准确性，并将源分离确定为主要瓶颈和未来研究的关键方向。
英文摘要
The inclusion of generative audio in the music production process has led to an increase in hybrid music tracks that blend authentic human performances with AI-generated stems, challenging traditional AI music detectors which operate in a binary setting. In this work, we propose a stem-agnostic framework for identifying synthetic audio sources within hybrid musical mixtures. We introduce the inspectrogram, a novel time-frequency representation that maps localized probabilities of synthetic content across the audio spectrum. By combining the inspectrogram with a Wiener filter estimating target stem energy dominance, a single CNN model evaluates whether the specific stem is generated. Trained on rendered hybrid mixtures and evaluated across various stem classes, our model achieves strong performance on high-frequency sources such as vocals, drums, and guitar, but struggles on the low-frequency, narrow-band bass. We conclude that the quality of separation impacts the detection accuracy and identify source separation as a primary bottleneck and a crucial direction for future research.
Comments4 pages + references, 5 figures, submitted to the 2027 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)
