MP3 to MIDI guitar chord check: melody and simultaneous notes are different tests.
In this controlled run, the guitar-like melody produced five events, while the guitar-like chord produced no clear notes under the same detector setup.
Input condition
Generated mono PCM WAV signals used a decaying guitar-like harmonic envelope. The melody fixture contained pitches 64, 67, 69, 71, and 74; the chord fixture contained pitches 52, 55, and 59. Both were evaluated with the deployed detector and MIDI writer in Node VM.
| Input | Observed output | What to inspect |
|---|---|---|
| guitar-like-melody.wav | Five events: pitches 64, 67, 69, 71, 74 | Sequential melody boundaries |
| guitar-like-chord.wav | Zero clear notes | Whether simultaneous tones are detected |
Outputs: guitar-like-melody.mid, guitar-like-chord.mid, and results.json.
What this does not prove
This comparison does not establish guitar transcription accuracy, polyphonic coverage, or behavior on recorded performances. It documents only these generated melody and chord fixtures.
After conversion
Separate a melody check from a chord check, inspect the generated MIDI, and do not treat a successful melody run as proof that simultaneous notes will be recovered.
Questions about guitar-like input
Did the melody and chord fixtures behave the same?
No. The melody produced five events, while the chord fixture produced no clear notes.
Does a melody result prove chord recovery?
No. They are different input conditions and must be checked separately.
What should I keep with the result?
Keep the exact input WAV, MIDI output, and results record together.