Tracing Inputs, Verifying Outputs: Validating Attribution in Music Generation

Key points

• 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.1
• 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.1
• We therefore audit memorization with our musical version identification model, musicDNA, and find few reproductions outside the input records.1
• 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.1

Sources (1)

• [1]Tracing Inputs, Verifying Outputs: Validating Attribution in Music Generation
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 7, 08:14 AM

  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.

Extractive summary: sentences quoted from the sources.