Fusion Segment Transformer: Bi-Directional Attention Guided Fusion Network for AI-Generated Music Detection

With the rise of generative AI technology, anyone can now easily create and deploy AI-generated music, which has heightened the need for technical solutions to address copyright and ownership issues. While existing works mainly focused on short-audio, the challenge of full-audio detection, which requires modeling long-term structure and context, remains insufficiently explored. To address this, we propose an improved version of the Segment Transformer, termed the Fusion Segment Transformer. As in our previous work, we extract content embeddings from short music segments using diverse feature extractors. Furthermore, we enhance the architecture for full-audio AI-generated music detection by introducing a Gated Fusion Layer that effectively integrates content and structural information, enabling the capture of long-term context. Experiments on the SONICS and AIME datasets show that our approach outperforms the previous model and recent baselines, achieving state-of-the-art results in AI-generated music detection.

Key Contributions

• Fusion Segment Transformer extending prior Segment Transformer with a Gated Fusion Layer combining content and structural streams via bi-directional cross-attention
• Integration of Muffin Encoder into the Stage-1 embedding pipeline to capture high-frequency spectral artifacts across multi-band Mel-spectrograms
• State-of-the-art AI-generated music detection results on SONICS and AIME datasets for full-length audio

🛡️ Threat Analysis

Details

Similar Papers