Get Started: Verifying Sheet Music Transcriptions with AI OCR

Optical character recognition (OCR) of sheet music often outputs typos (e.g. “Cmaj7” read as “Cnaj7”). This verifier checks chord legality.

By querying the /v2/resolve endpoint, it automatically detects malformed chords and flags them for manual correction.

Legality Check:

async function verifyChordSymbol(symbol) {
  try {
    await thiriClient.post('/v2/resolve', { chord: symbol });
    return true; // Chord is valid
  } catch {
    return false; // Typo detected
  }
}

🤖 File an Issue for an Autonomous Coding Agent

This project requires setting up an external application or workflow environment. If you want an autonomous coding agent (like Claude Code) to implement it in your repository:

1. Click here to open a pre-filled GitHub Issue on the repository
2. Submit the issue to trigger your repository’s autonomous builder agent.

Alternatively, you can copy the raw agent prompt instruction below:

Task Context: Build a code project utilizing the THIRI Model Context Protocol (MCP) server. Prompt Instruction: Write a Python correction script. The script should ingest text outputs from OCR models scanning music sheets, isolate suspected chord strings, and run them against the THIRI MCP resolve_chord tool. If the server throws an error (indicating a typo), suggest spelling corrections using Levenshtein distance matching against legal chord symbols.