BotCurrent↗
← Back to examples

Community recipe · Tool workflows

Reconstruct photographed sheet music as an editable score

Source-backed example. Practical decision guidance is proposed and untested; this is not a full implementation guide.

The problem

Photographed scores cannot be edited directly, and recognition errors can distort notes, rhythm and instrument alignment.

Agent job

Compare optical-music-recognition outputs, check structure, repair suitable errors and export an editable score with previews.

Expected output

Reviewable MusicXML and rendered score previews.

Where this framing comes from
  • Core problem: Editorial purpose framing
  • Agent job: Editorial purpose framing
  • Expected output: Editorial purpose framing
  • Relevant job roles: Editorial purpose framing
  • Benefit type: Editorial purpose framing
Editorial framing and outcome limits

Describes the task’s purpose and intended output; measured benefit, safe unsupervised operation and successful delivery are not established by this framing.

Editorial synthesis based on the linked source. This framing does not add evidence of successful delivery or results.

Case-specific decision guidance · untested

Case-specific decision guidance and first checks are original proposals. The linked source supports the attributed task; this added guidance has not been executed.

Read the attributed source evidence and original caveats below ↓

Why use this

  1. A musician with photographed notation may want to transpose, annotate or rearrange a score, yet recognition errors can be difficult to spot in a plausible-looking export. Several pieces on one page add another opportunity to merge unrelated music.

  2. This job could prepare an editable reconstruction while preserving the uncertain bars for musical review. Comparing structure and rendered notation gives the musician a focused way to inspect changes before relying on the score in rehearsal.

Why an agent / simpler alternative

Useful judgment

Compare recognition candidates, locate structural inconsistencies and retain repairs only when they improve the relevant passage.

Simpler alternative

Enter a short or especially unclear passage manually in notation software.

Unsuitable when

The source is illegible, rights are unclear or no musician can verify pitches and rhythm.

Proposed inputs

  1. Name

    Authorized score images

    Owner

    Musician

    Minimum scope

    Legible pages with piece boundaries and page order

    Quality check

    Check cropping, staff visibility and source rights

  2. Name

    Review brief

    Owner

    Musician

    Minimum scope

    Instruments, known meter/key and intended editing use

    Quality check

    Treat uncertain metadata as unknown rather than forcing it

Access and setup — proposed or unknown

Proposed minimum
  1. Copies of source images and isolated output files; no software installation or external upload in the first check.

Compatibility unknowns
  1. Recognition-engine versions, notation support, renderer availability and current runtime compatibility are unknown.

External actions requiring approval
  1. Downloads, installations, uploads and sharing reconstructed scores require appropriate approval.

Expected deliverable and review checks

Proposed artifact

MusicXML candidate, rendered previews and a bar-level uncertainty/change list.

Critical checks
  1. Compare pitches, accidentals, durations, meter and staff alignment; verify piece boundaries and export readability.

Missing output evidence

No inspected MusicXML file, recognition benchmark or musician-approved reconstruction is supplied here.

Failure signals and proposed responses

  1. Signal

    Measures balance but contain wrong pitches

    Why it matters

    Structural validity can hide musical errors

    Proposed response

    Compare the affected notes visually with the source.

  2. Signal

    A repair fixes one voice but shifts another

    Why it matters

    Local correction can damage ensemble alignment

    Proposed response

    Revert the repair and review the full measure together.

Effort and cost — unknown

Setup work
  1. Prepare images and choose notation review criteria.

Ongoing review
  1. Musician reviews uncertain bars and compares repairs.

Unknowns
  1. Recognition coverage, dependency setup, manual correction effort and costs are unknown.

First check — proposed, unexecuted

Status

proposed_unexecuted

Synthetic fixture

Specify an original fictional four-bar score with two voices, one accidental and a deliberate duration error in a candidate transcription; review the discrepancy list only.

Acceptance checks
  1. Flag duration and accidental checks; inspect both voices; require a preview comparison before accepting a repair.

Stop conditions
  1. Stop before installation or when the image cannot resolve an ambiguous symbol.

Limits

  1. A clean rendering or valid MusicXML does not demonstrate that the music matches the source.

Tasks: Audio & videoResearch & discovery

Useful for: MusicianEducator

Editorial relevance from the title and source-supported actions; not evidence of the operator’s occupation.
Claude CodeIndependent testing: No

Editorial candidate. This is source-backed reporting or a documented workflow. BotCurrent has not independently tested this task. A proposed Canadian adaptation is not evidence of a Canadian deployment or compliance certification.

Original source

https://github.com/Neanderthal/omr-notes/blob/main/skills/omr-notes/SKILL.md ↗

Attributed to: Sergey Istomin (Neanderthal)

Community recipe
Original action type
tool_action
What the source supports
Splits multi-piece pages, compares recognition configurations, checks score structure, retains beneficial bar repairs and exports editable MusicXML with rendered previews.
Source outcome limits
Recipe for a reviewable score; recognition and repair can still be wrong.
Evidence attribution caveat
This community recipe describes a procedure; it does not prove successful execution or independently reproduced results.
Source approval description
Not specified.
Named tools
Claude Code; Audiveris; oemer; verovio; MuseScore 4; Python; optional TrOMR
Model or version
Unknown
Publication date
Unknown
Source last checked
2026-10-14
Demonstrated in Canada
Unknown
Canadian availability
Unknown
Proposed Canadian adaptation
Use authorized Canadian repertoire or public-domain material; check page size and notation labels.
Recommended approval boundaries
Approve downloads and source rights; musician reviews pitches, rhythm and meter before use.