发表机构
Neutune; KAIST（Neutune; 韩国科学技术院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出通过输入追踪与输出验证相结合的方法，为AI音乐生成提供可验证的归因证据，并利用MixAudio和musicDNA模型验证了该方法在归因与记忆化检测上的有效性。
AI 中文摘要
我们如何验证AI生成的输出中包含了谁的音乐贡献？本文展示了基于输入的归因如何提供可验证的证据，证明生成过程中使用了哪些音频源，以及它们是否影响了输出。为此，我们仅基于音频进行生成，不涉及任何文本输入，然后追踪每个输出背后的输入，并确定其音乐效果。在提示遵循测试和受控输入交换中，我们的生成器MixAudio生成的音轨在音色上遵循提示音频，在和声上遵循上下文音频。然而，这些输出仍可能复制未作为输入提供的训练数据。因此，我们使用我们的音乐版本识别模型musicDNA审计记忆化，发现输入记录之外的复制情况很少。在被标记的池中的人工评判案例中，它比其他测试的记忆化检测器实现了更高的精确率和召回率。这两项评估表明，输入记录和输出分析为归因提供了互补的证据，随着AI音乐经济的形成，权利持有人的报告和补偿可以依据这些证据。音频示例可在此https URL获取。
英文摘要
How can we verify whose music contributed to an AI-generated output? This paper demonstrates how input-based attribution can provide verifiable evidence of which audio sources were used in a generation and whether they shaped the output. To do so, we condition the generation solely on audio without any text input, then trace the inputs behind each output, and establish their musical effect. In prompt adherence tests and controlled input swaps, the stems generated by our generator, MixAudio, follow the prompt audio in timbre and the context audio in harmony. Yet these outputs may still reproduce training data not supplied as inputs. We therefore audit memorization with our musical version identification model, musicDNA, and find few reproductions outside the input records. On human-judged cases within the flagged pool, it achieves higher precision and recall than the other tested memorization detectors. The two evaluations suggest that input records and output analysis provide complementary evidence for attribution, on which rights-holder reporting and compensation can draw as the AI music economy takes shape. Audio examples are available at https://neutune.github.io/attr2027demo/
Comments15 pages, 4 figures. Audio examples: https://neutune.github.io/attr2027demo/
