Audio to MIDI

The tool accepts an audio file and outputs a Standard MIDI File, enabling programmatic access to pitch events and timing data. It supports mono and polyphonic transcription depending on the selected mode.
Conceptually the pipeline decodes audio, extracts pitch tracks, detects onsets, quantizes timing, and maps events to MIDI note-on and note-off with velocities. It then exports a MIDI file with a fixed ticks-per-quarter-note resolution.
Target users include composers integrating transcription into DAWs, game audio teams generating background music, researchers building datasets, and developers embedding MIDI generation in Python workflows.
Unique value lies in its mode options, optional tempo estimation, batch processing capability, and a straightforward backend suitable for automation. Use cases include transcribing riffs, vocal lines, or orchestral phrases for scoring experiments or rapid prototyping.