Recognition with evidence
StaveWave treats sheet music OCR as a reviewable workflow, keeping engine votes, validation findings, quality checks, and weak spots visible.
Sheet music OCR
Turn scanned sheet music into editable MusicXML and see the OMR votes, specialist checks, and confidence notes that explain the result.
Optimized for sheet music OCR, optical music recognition, OMR software, music OCR workflows with reviewable MusicXML output.

StaveWave treats sheet music OCR as a reviewable workflow, keeping engine votes, validation findings, quality checks, and weak spots visible.
Instead of hiding uncertainty, the conversion report points to measures where engines or validators found risk.
Export MusicXML for editing, arranging, cleanup, and import into common score-writing applications.
Processing stack
StaveWave shows the recognition engines, specialist passes, validation checks, and quality gates involved so users can see what checked the score before trusting the MusicXML.
OMR engines
Audiveris, homr, and oemer each try the notation independently. StaveWave compares their outputs and keeps disagreements visible instead of treating one reading as automatically correct.
Review checks
Visual, corpus, and notation checks inspect weak spots, compare source evidence against rendered MusicXML, and flag suspicious measures or repair candidates.
Specialist checks
Focused passes review pitch, rhythm, text, structure, articulation, and style so the output is easier to clean up in Sibelius, Dorico, MuseScore, Finale, or another editor.
Quality gates
Validation, consensus, visual comparison, audio checks, plausibility scoring, and audit evidence must run. If a required component fails, the job reports a blocking error.
Workflow
Use a PDF, PNG, JPG, WEBP, or TIFF score file up to 50 MB and 20 pages.
Review the status page as file checks, recognition engines, validation, and quality gates run.
Use the review report to focus review time on measures with warnings or lower routing scores.
Questions
These answers match the structured data search engines receive for this page.
No. Sheet music OCR, also called optical music recognition, reads notation symbols and exports structured music data.
StaveWave accepts PDF, PNG, JPG, WEBP, and TIFF files up to 50 MB and 20 pages.
Music notation mistakes can be subtle. Review notes help you find measures that deserve a closer look.
Ready to convert