Skip to content

Music Attribution Scaffold

Open-source research scaffold for music attribution with transparent confidence scoring.

Companion code to: Teikari, P. (2026). Governing Generative Music: Attribution Limits, Platform Incentives, and the Future of Creator Income. SSRN No. 6109087.

Music Attribution Scaffold: editorial frontend with confidence gauges, assurance badges, review queue, and agentic sidebar for transparent music credit verification.

What is This?

Music Attribution Scaffold is a research framework that demonstrates how to build a multi-source music attribution system with:

  • Per-field confidence scores — not just "is this correct?" but "how confident are we, and why?"
  • Entity resolution across messy sources — reconciling MusicBrainz, Discogs, AcoustID, and file metadata
  • AI-ready permissions via MCP — machine-readable consent infrastructure for the generative AI era

Research Scaffold, Not Production System

This repo demonstrates concepts from the SSRN preprint. It is designed to be forked, extended, and adapted — not deployed as-is.

The Pipeline

Five-pipeline architecture: ETL ingests from 5 sources, entity resolution reconciles identities, attribution engine scores confidence, API/MCP serves data, chat agent enables verification.

The five-pipeline architecture: data flows from ETL through entity resolution and attribution scoring to API/MCP endpoints and the agentic chat interface.

Pipeline diagram (Mermaid) 
graph LR
    A[ETL<br/>5 sources] --> B[Entity<br/>Resolution]
    B --> C[Attribution<br/>Engine]
    C --> D[API / MCP<br/>Server]
    D --> E[Chat Agent<br/>+ Frontend]

    style A fill:#e1f5fe
    style B fill:#e1f5fe
    style C fill:#fff3e0
    style D fill:#e8f5e9
    style E fill:#e8f5e9

Quick Start

git clone https://github.com/petteriTeikari/music-attribution-scaffold.git
cd music-attribution-scaffold
make setup        # Install deps + start PostgreSQL
make agent &      # Start FastAPI backend on :8000
make dev-frontend # Start Next.js frontend on :3000

Open http://localhost:3000 to see 9 Imogen Heap attribution records with confidence scoring.

  • Getting Started

    Install and run the scaffold in under 5 minutes

    Installation

  • Key Concepts

    Understand the Oracle Problem, assurance levels, and two-friction taxonomy

    Concepts

  • API Reference

    Auto-generated documentation from source code

    API Docs

  • Reproduce the Paper

    Map every paper claim to code, tests, and demo commands

    Paper Guide

Sample Data

The scaffold ships with 9 Imogen Heap works spanning the full confidence range:

Work Confidence Assurance Needs Review
Hide and Seek 0.95 A3 No
Tiny Human 0.91 A3 No
The Moment I Said It 0.82 A2 No
Goodnight and Go 0.72 A2 No
Headlock 0.58 A1 Yes
What Have You Done To Me? 0.48 A1 Yes
Just for Now 0.35 A1 Yes
2-1 0.28 A1 Yes
Blanket 0.00 A0 Yes

Technology Stack

Layer Technology Why
Backend Python 3.13, FastAPI, SQLAlchemy Async-first, type-safe, Pydantic integration
Database PostgreSQL + pgvector Relational + vector search in one database
Agent PydanticAI, AG-UI protocol Type-safe tools, SSE streaming, FallbackModel
Frontend Next.js 15, Tailwind v4, CopilotKit App Router, editorial design system
State Jotai Atomic state, no boilerplate
Analytics PostHog Privacy-first, typed events
MCP Model Context Protocol Machine-readable permission queries

Citation

@article{teikariGoverningGenerativeMusic2026,
    title = {Governing Generative Music: Attribution Limits, Platform Incentives, and the Future of Creator Income},
    url = {https://doi.org/10.2139/ssrn.6109087},
    doi = {10.2139/ssrn.6109087},
    publisher = {Social Science Research Network},
    author = {Teikari, Petteri},
    year = {2026},
}

License

MIT. All dependencies verified MIT/Apache/BSD compatible.