compareMusic

An evaluation function for the Lambda Feedback platform. It compares a student's music performance against a reference performance and returns formative feedback on pitch accuracy, timing, note duration and chords.

The student's response and the reference answer can each be MIDI note data, a path or URL to a MIDI file (.mid / .midi), or a path or URL to an audio recording (.mp3, .wav, .ogg, .aac, .flac, .opus or .m4a), which is transcribed to MIDI before the comparison.

Deployment

Documentation

• docs/user.md — for teachers setting up a question: input format, what the student sees, and the parameters that control strictness.
• docs/dev.md — inputs, outputs and worked examples.

Both files are published to the Lambda Feedback documentation site.

Repository structure

evaluation_function/
    compare_MIDI.py       # alignment, scoring and feedback generation
    audio_processing.py   # audio-to-MIDI transcription (Basic Pitch)
    evaluation.py         # platform entry point, thin wrapper
    preview.py            # preview entry point
    main.py               # server entry point
    *_test.py             # tests

data/                     # fixtures used by the tests
docs/                     # user and developer documentation
notebooks/                # development and evaluation notebooks, see notebooks/README.md
config.json               # evaluation function name used when deploying

Working on this repository

poetry install
poetry run pytest

To run the function in a container:

docker build -t compare-music .
docker run --rm -p 9000:8080 compare-music

It then answers POST http://localhost:9000/ with a command: eval header and a body of {"response": ..., "answer": ..., "params": {}}. GET /health reports readiness.