Назад к тому, как работает распознаваниеЧто нового в OMR

Сканирование PDF и фото: что нового.

Сканирование в Opuscan на базе оптического распознавания нот (OMR) становится все лучше. На этой странице собраны обновления импорта PDF и фотографий, начиная с самых новых.

Recent improvements

  • NewCourtesy (cautionary) accidentals are now recognized.
  • FixFixed some multi-voice passages where the two voices share a rest.

Sharper page analysis for complex layouts

We've improved how scanning reads the layout of a page. Imports of PDFs and photos now produce cleaner, more accurate scores, especially on complex layouts:

  • BetterDense pages now split correctly. Pages with many short systems stacked together are detected separately, no longer merged.
  • BetterOrchestral and large-ensemble scores. Multi-staff instruments like piano and organ, and scores with many staves, are recognized more reliably.
  • BetterWide, landscape, and booklet layouts. Landscape pages no longer clip the left edge, and multi-page and booklet scans are handled correctly.
  • FixFewer stray detections. Spurious elements near page footers and margins are no longer picked up as music.

Notation and recognition improvements

  • NewCustom beaming patterns are now recognized and preserved on import.
  • NewBeams that span rests (note / rest / note) are now supported.
  • NewDetected jazz scores apply a jazz chord style and handwritten jazz music font automatically.

A more accurate model

Every PDF and photo import is now processed by Tutteo's own in-house model. This release is a major new version focused on transcription accuracy - musical correctness improves across every test set, with the largest gains in:

Notation Improvement
Dynamics +39%
Fermatas, articulations, ties, accidentals +30 to 37%
Note durations +22%
Slurs, dotted notes, rests, staff & voice assignment +14 to 20%

Text, dynamics, and layout placement are also more precise: lyrics, chord names, and dynamics attach more reliably to the correct notes, including on complex layouts like piano grand staves and parts with clef changes.

Our own AI model starts rolling out

We're rolling out Tutteo's own AI model for PDF and photo import, designed and trained in-house to replace the third-party engine used at launch. Owning the technology end to end lets us push accuracy further, broaden what we recognize, ship improvements faster, and align results with Tutteo's own notation format. This first release already reads much more of your music:

Broader notation coverage - newly recognized:

  • NewChord symbols, encoded with their musical meaning as editable chords - not just plain text
  • NewMeasure repeats and rhythm slashes
  • NewTempo and metronome markings (with BPM)

Lyrics and text in more languages. Lyrics, rehearsal marks, and other text in Latin-script languages by default, plus Japanese, Korean, and Chinese.

Truer to the original layout. Imports follow the source score's layout much more closely - measures per line, systems per page, and overall structure - making a real difference on large, complex scores and supporting more page formats, including landscape.