Sheet Music Publisher

Your Task

Input: $ARGUMENTS

Guide user through sheet music generation from mastered audio:

1. Setup verification - Check AnthemScore and MuseScore installed 2. Track selection - Identify which tracks to transcribe (melodic tracks work best) 3. Automated transcription - Run transcribe.py via AnthemScore CLI 4. Optional polish - Recommend MuseScore editing for accuracy improvements 5. Prepare singles - Create clean-titled consumer-ready files (PDF, XML, MIDI) 6. Optional songbook - Create distribution-ready combined PDF with TOC

External Software Requirements

REQUIRED:

• AnthemScore ($42 Professional edition) - Audio transcription engine
• Free trial: 30 seconds per song, 100 total transcriptions
• Download: https://www.lunaverus.com/
• Cross-platform: macOS, Linux, Windows

• MuseScore (Free) - Notation editing and PDF export
• Download: https://musescore.org/
• Cross-platform: macOS, Linux, Windows

Python dependencies (songbook only):

pip install pypdf reportlab pyyaml

Check if user has these installed FIRST before proceeding.

Supporting Files

• anthemscore-reference.md - AnthemScore CLI reference, installation
• musescore-reference.md - MuseScore polish techniques
• publishing-guide.md - Distribution guide, licensing considerations
• ../../reference/sheet-music/workflow.md - Complete workflow documentation
• workflow-detail.md - Detailed workflow phases, error handling, tips, tool examples

---

Sheet Music Publisher Agent

You are a sheet music production specialist. Your role is to guide users through converting mastered audio into publishing-quality sheet music and songbooks.

Core Responsibilities

1. Setup verification - Ensure required software installed 2. Track triage - Identify suitable candidates for transcription 3. Automated batch processing - Use AnthemScore CLI for efficiency 4. Quality control - Recommend polish where needed 5. Publication preparation - Prepare singles and distribution-ready songbooks

Understanding the User's Context

Resolve paths via MCP: 1. Call get_config() — returns audio_root, content_root, artist.name 2. Call find_album(album_name) — fuzzy match to get album slug and metadata 3. Call resolve_path("audio", album_slug) — returns the audio directory path

Sheet music output:

{audio_path}/sheet-music/
├── source/        # AnthemScore output (numbered files)
├── singles/       # Consumer-ready downloads (clean titles, all formats)
│   └── .manifest.json
└── songbook/      # Combined songbook PDF

---

Override Support

Check for custom sheet music preferences:

Loading Override

1. Call load_override("sheet-music-preferences.md") — returns override content if found (auto-resolves path from config) 2. If found: read and incorporate preferences 3. If not found: use base sheet music workflow only

Override File Format

`{overrides}/sheet-music-preferences.md`:

# Sheet Music Preferences

## Page Layout
- Page size: letter (8.5x11) or 9x12 (standard songbook)
- Margins: 0.5" all sides (override: 0.75" for wider pages)
- Font: Bravura (default) or MuseJazz for jazz albums
- Staff size: 7mm (default) or 8mm for large print

## Title Formatting
- Include track numbers: no (default) or yes
- Title position: centered (default) or left-aligned
- Composer credit: "Music by [artist]" below title
- Copyright notice: © 2026 [artist]. All rights reserved.

## Notation Preferences
- Clefs: Treble and bass (piano) or single staff (melody only)
- Key signatures: Shown (default) or omitted for atonal music
- Time signatures: Shown (default) or omitted for free time
- Tempo markings: BPM numbers or Italian terms

## Songbook Settings
- Table of contents: yes (default) or no
- Page numbers: bottom center (default) or bottom right
- Section headers: by genre (default) or chronological
- Cover page style: minimalist (title + artist) or elaborate (artwork)

## Transcription Settings
- Accuracy target: 85% (default) or 95% (requires manual polish)
- Polish level: minimal (quick) or detailed (time-consuming)
- Instrument focus: piano (default), guitar, or vocal melody
- Complexity: simplified (easier to play) or exact (harder, more accurate)

How to Use Override

1. Load at invocation start 2. Apply page layout preferences to songbook creation 3. Use title formatting rules consistently 4. Follow notation preferences when polishing 5. Apply songbook settings to combined PDF 6. Override preferences guide but don't compromise quality

Example:

• User prefers 9x12 page size, large print
• User wants track numbers in titles
• Result: Generate songbook with 9x12 pages, 8mm staff, titles include track numbers

---

Workflow Phases

See workflow-detail.md for detailed steps on all 7 phases:

1. Setup Verification (AnthemScore, MuseScore, Python deps) 2. Track Selection 3. Automated Transcription (outputs to source/) 4. Quality Review & Polish 5. Prepare Singles (clean titles → singles/) 6. Songbook Creation (optional → songbook/) 7. Summary & Next Steps

Also covers: Error Handling, Tips for Better Results, Tool Invocation Examples, Quality Standards, Workflow State Tracking.

Remember

1. Load override first - Call load_override("sheet-music-preferences.md") at invocation 2. Apply formatting preferences - Use override page layout, notation, songbook settings if available 3. Use MCP for paths - Call get_config(), find_album(), resolve_path("audio") instead of reading config manually 4. Check software exists - Graceful failure with install instructions 5. Set expectations - 70-95% accuracy, may need polish 6. Offer polish - Don't skip this step 7. Automate what you can - Use CLI tools, minimize manual work 8. distribution-ready output - Songbook should be upload-ready (with user preferences applied)

Success Criteria

User should end with:

• ✓ Individual PDFs for each track (publishing-ready)
• ✓ MusicXML sources (editable in MuseScore)
• ✓ MIDI files for each track (playback)
• ✓ Optional: Combined songbook PDF (distribution-ready)
• ✓ Clear next steps for website distribution
• ✓ Understanding of quality level and polish needs

AnthemScore CLI Reference

Complete reference for AnthemScore command-line interface and setup.

Installation

Purchase & Download

Website: lunaverus.com

Editions:

Free trial:

• 30 seconds per song
• 100 total transcriptions
• All features available
• No subscription needed

Platform Support

• Windows (7, 8, 10, 11)
• macOS (10.13+)
• Linux (Ubuntu, Fedora, Arch)

No subscription - Buy once, use forever

---

CLI Executable Locations

macOS

/Applications/AnthemScore.app/Contents/MacOS/AnthemScore

Verify installation:

ls -l "/Applications/AnthemScore.app/Contents/MacOS/AnthemScore"

Linux

/usr/bin/anthemscore
/usr/local/bin/anthemscore

Find via PATH:

which anthemscore

Windows

C:\Program Files\AnthemScore\AnthemScore.exe
C:\Program Files (x86)\AnthemScore\AnthemScore.exe

---

Basic CLI Usage

Headless Mode (No GUI)

Required flag: -a or --headless

Without this flag, AnthemScore opens GUI. Use -a for batch automation.

Simple Examples

WAV → PDF only:

AnthemScore track.wav -a -p output.pdf

WAV → PDF + MusicXML:

AnthemScore track.wav -a -p output.pdf -x output.xml

WAV → PDF + MusicXML + MIDI:

AnthemScore track.wav -a -p output.pdf -x output.xml -m output.mid

---

All CLI Flags

Output Formats

Transcription Options

Piano Key Range

Piano key numbers:

• 1 = A0 (lowest key)
• 40 = Middle C
• 88 = C8 (highest key)

Audio Processing

Advanced

---

Common Use Cases

Batch Process Directory

#!/bin/bash
ANTHEMSCORE="/Applications/AnthemScore.app/Contents/MacOS/AnthemScore"

for wav in *.wav; do
  basename="${wav%.wav}"
  echo "Processing: $wav"
  "$ANTHEMSCORE" "$wav" -a -p "$basename.pdf" -x "$basename.xml"
done

Treble Clef Only (Vocal Melody)

AnthemScore vocal.wav -a -t -p vocal.pdf

Use when:

• Transcribing vocal melodies only
• High-pitched instruments (flute, violin)
• Simpler output with one staff

Bass Clef Only

AnthemScore bass.wav -a -b -p bass.pdf

Use when:

• Transcribing bass lines only
• Low-pitched instruments (cello, bass guitar)

Limit to Piano Range

# Middle C (40) to C6 (76)
AnthemScore track.wav -a -j 40 -k 76 -r -p track.pdf

Use when:

• Avoiding ghost notes in extreme ranges
• Focusing on playable piano range

Export MusicXML Only (No PDF)

AnthemScore track.wav -a -x track.xml

Use when:

• Planning to edit in MuseScore and export PDF there
• Faster processing (skip PDF rendering)

Save Project for GUI Editing

AnthemScore track.wav -a -d track.asdt

Then open in GUI:

• Adjust sensitivity slider
• Manually correct errors
• Re-export

---

Transcription Quality

Accuracy Expectations

Factors affecting accuracy:

• Audio quality (clear > muddy)
• Arrangement density (simple > complex)
• Frequency range (mid-range > extreme)
• Recording quality (studio > live)

What Transcribes Well

✓ Clear vocal melodies ✓ Simple accompaniment ✓ Acoustic instruments ✓ Piano-based songs ✓ Singer-songwriter ✓ Folk, country, ballads

What's Challenging

⚠ Polyphonic harmony ⚠ Dense arrangements ⚠ Heavy distortion ⚠ Rapid percussion ⚠ Electronic synth pads ⚠ Extreme frequency content

---

Performance

Processing Time

Typical: 30-60 seconds per track (3-5 minute song)

Factors:

• CPU speed (single-threaded mostly)
• Audio length
• Complexity (dense = slower)
• Output formats (PDF + XML + MIDI = longer)

Speed Tips

Faster:

• Use -p only (skip MusicXML/MIDI if not needed)
• Limit key range with -j/-k flags
• Increase threads with -z flag

Slower (but better):

• Export all formats for flexibility
• Full piano range (no limits)
• Save project .asdt for later editing

---

Troubleshooting

"License activation failed"

Solution: Run GUI once to activate license

# macOS - open GUI
open -a AnthemScore

# After activation, CLI will work

"File not found" error

Check:

• Full path to AnthemScore executable correct
• WAV file exists and readable
• Output directory exists (create with mkdir -p)

"Transcription produced no notes"

Possible causes:

• Note sensitivity too low (can't adjust in CLI)
• Audio file is silent or very quiet
• File format not supported (use WAV, not MP3)

Solution: Open in GUI, adjust sensitivity slider, re-export

Very slow processing

Check:

• File is reasonable length (not 30-minute track)
• System isn't under heavy load
• Disk space available

---

Best Practices

File Naming

Good:

01-ocean-of-tears.wav → 01-ocean-of-tears.pdf
track-name.wav → track-name.xml

Bad:

track 1.wav        # Spaces require quoting
FINAL-v2-MASTER.wav # Too long, unclear

Output Organization

album/
├── mastered/
│   ├── 01-track.wav
│   └── 02-track.wav
└── sheet-music/
    ├── 01-track.pdf
    ├── 01-track.xml
    └── 02-track.pdf

Keep source WAVs separate from sheet music output.

Batch Processing

Do:

• Process entire album at once
• Log output for debugging
• Use consistent naming

Don't:

• Process in GUI one-by-one (slow)
• Mix source and output in same folder
• Forget to save MusicXML for editing

Quality Control

After batch processing: 1. Review PDFs for obvious errors 2. Check key signatures (AI detection isn't perfect) 3. Verify playback with MIDI files 4. Identify tracks needing manual polish 5. Edit MusicXML in MuseScore, re-export PDFs

---

Integration with MuseScore

Workflow

AnthemScore CLI → MusicXML → MuseScore → Polished PDF

Why this workflow:

• AnthemScore: Fast automated transcription
• MuseScore: Manual corrections and formatting
• Final PDF: Publishing-ready output

MusicXML Benefits

• Preserves notation data (notes, rests, timing)
• Editable in any notation software
• Better than PDF for corrections
• Smaller file size than .asdt projects

When to Skip MuseScore

Skip if:

• PDFs are "good enough" for reference
• Quick transcription for personal use
• Planning to transcribe by ear anyway

Don't skip if:

• Publishing to KDP
• Professional licensing
• Teaching materials
• Accuracy matters

---

Advanced Techniques

Stem Separation + Transcription

For complex arrangements:

# 1. Separate stems with Demucs
demucs track.wav

# 2. Transcribe each stem
AnthemScore separated/track/vocals.wav -a -p vocals.pdf -x vocals.xml
AnthemScore separated/track/bass.wav -a -p bass.pdf -x bass.xml

# 3. Combine in MuseScore manually

Benefits:

• Higher accuracy for dense mixes
• Separate melody from harmony
• Better bass line detection

Limiting Range to Reduce Noise

# Remove very high/low notes that are likely errors
AnthemScore track.wav -a -j 21 -k 76 -r -p track.pdf

Piano key reference:

• 21 = A1 (low A on piano)
• 76 = C6 (two octaves above middle C)

Custom Note Resolution

# Don't transcribe notes shorter than eighth notes
AnthemScore track.wav -a -w 8 -p track.pdf

Values:

• -w 4 = Quarter notes (slow ballads)
• -w 8 = Eighth notes (most songs)
• -w 16 = Sixteenth notes (fast/detailed)

---

Reference: Piano Key Numbers

Typical vocal ranges:

• Soprano: 48-79 (C4-G5)
• Alto: 43-74 (G3-D5)
• Tenor: 40-69 (C4-A4)
• Bass: 33-62 (A2-D4)

---

Getting Help

CLI help:

AnthemScore --help

Official documentation:

• Website: https://www.lunaverus.com/
• Support: support@lunaverus.com

Community:

• Discord server (link on website)
• YouTube tutorials
• User forums

MuseScore Polish Reference

Techniques for editing and polishing sheet music transcriptions in MuseScore.

Installation

Download & Install

Website: musescore.org

Cost: Free, open source

Supported platforms:

• Windows (7, 8, 10, 11)
• macOS (10.12+)
• Linux (AppImage, Ubuntu/Debian, Fedora, Arch)

Current versions:

• MuseScore 4 (latest, recommended)
• MuseScore 3 (stable, still supported)

No account required for basic use. Optional MuseScore.com account for cloud features.

---

Opening AnthemScore Output

Import MusicXML

File → Open

• Navigate to .xml or .musicxml file
• MuseScore auto-detects format
• Import settings usually correct (accept defaults)

First review:

• Play through (Space bar)
• Check key signature
• Check time signature
• Verify tempo marking

---

Essential Corrections

Key Signature

Problem: AI often guesses wrong, especially for modes or key changes

Fix: 1. Select measure where key change should occur 2. Edit → Key Signature (or press K) 3. Choose correct key from palette 4. Apply to: "All staves" or "Just this staff"

Tip: If fundamentally wrong key (e.g., C major instead of A minor), you may need to transpose the entire score.

Time Signature

Problem: AI can misinterpret syncopation as different time signature

Fix: 1. Select measure 2. Edit → Time Signature (or press T) 3. Choose correct time signature 4. Apply: "From this measure onward" or "Just this measure"

Common fixes:

• 4/4 mistaken for 2/4 (or vice versa)
• Swing feel transcribed as triplets in 4/4

Tempo Marking

Problem: Often missing or incorrect BPM

Fix: 1. Select first measure (or where tempo changes) 2. Add → Text → Tempo Marking 3. Type tempo (e.g., "♩ = 120" or "Moderate") 4. MuseScore auto-converts text to symbols

Standard terms:

• Largo (40-60 BPM)
• Adagio (60-80 BPM)
• Moderato (90-110 BPM)
• Allegro (120-140 BPM)

---

Note Corrections

Wrong Note Pitches

Fix individual notes: 1. Click note to select 2. Press Up/Down arrow to move by half-step 3. Press Ctrl/Cmd + Up/Down to move by octave

Fix chords: 1. Click one note in chord 2. Shift + click other notes to select all 3. Arrow keys move entire selection

Wrong Note Durations

Shortcuts (with note selected):

• 5 = Quarter note
• 6 = Eighth note
• 7 = Sixteenth note
• 4 = Half note
• 3 = Whole note
• . = Dotted note (toggle)

Visual guide:

• ♩ = Quarter note (5)
• ♪ = Eighth note (6)
• ♬ = Sixteenth note (7)
• 𝅗𝅥 = Half note (4)

Adding Missing Notes

Enter note input mode: 1. Press N (or click note input icon) 2. Click staff where note should go 3. Type letter name (A-G) 4. Use Up/Down arrows to adjust octave 5. Use number keys for duration (4-7) 6. Press Esc when done

Deleting Wrong Notes

1. Click note to select 2. Press Delete or Backspace 3. May need to add rests to complete measure

---

Rest Corrections

Incomplete Measures

Problem: Measure shows "duration error" (doesn't add up to time signature)

Fix: 1. Select measure 2. Tools → Regroup Rhythms (auto-fixes common issues) 3. Or manually add rests to complete

Add rests:

• Press 0 (zero) in note input mode
• Or: Delete a note, rest auto-inserted

Simplify Rest Notation

Problem: Multiple short rests instead of one long rest

Fix:

• Tools → Regroup Rhythms (auto-consolidates)
• Manually: Delete short rests, add longer rest

---

Formatting & Layout

Page Breaks

Add page break: 1. Select barline where break should occur 2. Add → Breaks & Spacers → Page Break 3. Or: Press Ctrl/Cmd + Return

Best practices:

• Break at natural pause (end of verse, chorus)
• Avoid awkward page turns (mid-phrase)

System Breaks (Line Breaks)

Add system break: 1. Select barline 2. Add → Breaks & Spacers → System Break 3. Or: Press Return

Use when:

• Separating sections (verse, chorus)
• Improving readability
• Controlling measures per line

Spacing

Increase/decrease space:

• Format → Style → Measure → Measure spacing
• Increase for crowded measures
• Decrease to fit more on page

Manual adjustments:

• Select measure
• Format → Stretch
• Adjust to taste

---

Performance Markings

Dynamics

Add dynamics: 1. Select note or measure 2. Add → Dynamics (or press Ctrl/Cmd + E) 3. Choose from palette: p, mp, mf, f, ff, crescendo, etc.

Common dynamics:

• pp = Very soft
• p = Soft
• mp = Medium soft
• mf = Medium loud
• f = Loud
• ff = Very loud

Hairpins (crescendo/diminuendo):

• Select notes
• Click crescendo < or diminuendo > in palette
• Drag to adjust length

Articulations

Add articulation: 1. Select note 2. Double-click articulation in palette

• Staccato (dot above note)
• Accent (>)
• Tenuto (line)
• Fermata (pause)

Shortcuts:

• . = Staccato
• Shift + S = Slur

Pedal Markings (Piano)

Add pedal line: 1. Select note where pedal starts 2. Lines → Pedal (from palette) 3. Drag line to where pedal ends 4. Displays: "Ped." ... "*" (release)

---

Lyrics (Optional)

Add lyrics: 1. Select first note of phrase 2. Press Ctrl/Cmd + L 3. Type syllable, press Space for next note 4. Press - (hyphen) for multi-syllable word 5. Esc when done

Verses:

• Ctrl/Cmd + L = Verse 1
• Add → Text → Lyrics → Verse 2 (etc.)

---

Cleanup Tasks

Beaming

Problem: Eighth notes beamed incorrectly (or not beamed)

Fix: 1. Select notes 2. Format → Beam → Choose beam style

• Beam together
• Break beam
• Auto

Best practice: Beam by beat in 4/4 time

Ties vs. Slurs

Tie: Connects same note across barline (held) Slur: Connects different notes (phrasing)

Add tie: 1. Select first note 2. Press + (plus key)

Add slur: 1. Select first note 2. Press S 3. Drag to last note

Remove Ghost Notes

Problem: AI transcribed noise as notes

Fix: 1. Play through, listen for wrong notes 2. Compare to original audio 3. Delete notes that aren't in original

---

Title & Credits

Score Properties

Edit → Score Properties

Fields:

• Work Title: Song name (no track number!)
• Composer: Artist name
• Lyricist: If different from composer
• Copyright: © Year Artist Name

These appear:

• On first page
• In PDF metadata
• In printed scores

Header/Footer

Format → Style → Header, Footer

Customize:

• Page numbers
• Title on every page
• Copyright notice

---

Export to PDF

Export Settings

File → Export → PDF

Settings:

• Resolution: 300 DPI (print quality)
• Page size: Letter (US) or A4 (international)
• All pages or range

File naming:

• Use clean title (no track numbers)
• E.g., ocean-of-tears.pdf not 01-ocean-of-tears.pdf

---

MuseScore CLI (For Automation)

Batch PDF Export

macOS:

/Applications/MuseScore\ 4.app/Contents/MacOS/mscore -o output.pdf input.xml

Linux:

mscore -o output.pdf input.xml

Batch process:

for xml in *.xml; do
  mscore -o "${xml%.xml}.pdf" "$xml"
done

Find MuseScore Executable

Common locations:

• macOS 4: /Applications/MuseScore 4.app/Contents/MacOS/mscore
• macOS 3: /Applications/MuseScore 3.app/Contents/MacOS/mscore
• Linux: /usr/bin/musescore or which mscore
• Windows: C:\Program Files\MuseScore 4\bin\MuseScore4.exe

---

Keyboard Shortcuts Reference

Navigation

• Space: Play/pause
• Home: Jump to start
• End: Jump to end
• Page Up/Down: Scroll score

Note Input

• N: Toggle note input mode
• A-G: Enter note by letter
• 0: Enter rest
• 4-7: Note duration (whole, half, quarter, eighth)
• . (dot): Toggle dotted note

Editing

• Ctrl/Cmd + C: Copy
• Ctrl/Cmd + V: Paste
• Ctrl/Cmd + X: Cut
• Delete: Delete selection
• Ctrl/Cmd + Z: Undo
• Ctrl/Cmd + Y: Redo

Pitch

• Up/Down arrows: Pitch by half-step
• Ctrl/Cmd + Up/Down: Pitch by octave

Articulations

• .: Staccato
• S: Slur
• +: Tie

Layout

• Return: System break
• Ctrl/Cmd + Return: Page break
• K: Key signature
• T: Time signature

---

Common Fixes Checklist

After importing from AnthemScore, check:

• [ ] Key signature correct
• [ ] Time signature correct
• [ ] Tempo marking added
• [ ] All measures complete (no duration errors)
• [ ] Note pitches accurate (compare to audio)
• [ ] Note durations correct
• [ ] Beaming appropriate
• [ ] Dynamics added (at minimum: p, mf, f)
• [ ] Articulations added (staccato, accents)
• [ ] Page breaks at sensible locations
• [ ] Title/composer in score properties (no track numbers!)
• [ ] Copyright notice added
• [ ] Playback sounds correct (Space bar test)

---

Quality Standards

Minimum Viable

• Correct notes (mostly)
• Readable layout
• Title and composer
• Basic structure

Publishing Ready

• Clean, professional notation
• Correct key/time signatures
• Dynamics and articulations
• Proper formatting
• Copyright notice
• Page turns at natural breaks
• Playable by intermediate pianist

---

Troubleshooting

"MuseScore won't open XML file"

Check:

• File extension is .xml or .musicxml (not .mid)
• File isn't corrupted (try opening in text editor - should see XML tags)
• Using MuseScore 3+ (older versions may not support MusicXML 3.0)

"Playback sounds wrong"

Check:

• Instrument assigned correctly (piano?)
• Soundfont loaded (MuseScore → Preferences → I/O)
• MIDI device configured

Fix:

• Don't trust playback 100% - compare to original audio
• Playback is for verification only

"Notes are in wrong clef"

Fix: 1. Right-click staff 2. Staff/Part Properties 3. Change clef to treble or bass 4. Apply

"Can't export PDF - error"

Check:

• Disk space available
• Output directory exists and writable
• MuseScore has permissions
• No special characters in filename

---

Learning Resources

Built-in handbook:

• Help → Handbook (comprehensive guide)

Video tutorials:

• MuseScore YouTube channel
• Search: "MuseScore piano tutorial"

Community:

• MuseScore.org forums
• Discord server
• Facebook groups

---

Best Practices

Workflow

1. Import MusicXML from AnthemScore 2. Play through - identify obvious errors 3. Fix structure - key sig, time sig, tempo 4. Correct notes - compare to original audio 5. Add markings - dynamics, articulations 6. Format - page breaks, spacing 7. Properties - title, composer, copyright 8. Export PDF - 300 DPI, clean filename 9. Save project - keep .mscz for future edits

Save Your Work

File → Save (or Ctrl/Cmd + S)

Save as `.mscz` format - This is MuseScore's native format, editable later.

Keep both:

• .mscz - Source file (editable)
• .pdf - Deliverable (publishing-ready)

Version Control

For significant changes:

• Save numbered versions: track-v1.mscz, track-v2.mscz
• Or use git for .mscz files (they're ZIP archives, version-friendly)

---

Integration with Sheet Music Workflow

Position in Workflow

WAV → AnthemScore → MusicXML → MuseScore → Final PDF
                                    ↑
                              You are here

When to Polish

Always polish if:

• Publishing to KDP or selling
• Professional licensing package
• Teaching materials

Maybe polish:

• Personal reference sheets
• Quick transcriptions for learning

Skip polish if:

• Just need rough notation
• Time-constrained
• Audio quality was poor (transcription too messy)

---

Quick Reference Card

Sheet Music Distribution Guide

Guide for distributing sheet music via your website and other platforms.

Overview

What you're distributing: Piano arrangements (sheet music) as individual singles and a combined songbook

Primary method: Free download from your website

File formats:

Alternative platforms (optional):

• Sheet Music Plus — digital marketplace
• Musicnotes — digital sheet music store
• Your own website — direct downloads (recommended)

---

Website Distribution

Recommended Setup

Host sheet music as free downloads on your website alongside the album.

Per-track downloads (from singles/ directory):

• PDF — for viewing/printing
• MusicXML — for editing
• MIDI — for playback

Complete songbook (from songbook/ directory):

• Single PDF with title page, copyright, TOC, all tracks

File Organization for Upload

The pipeline produces a clean directory structure:

sheet-music/
├── source/          # Working files (don't distribute)
├── singles/         # Upload these — one set per track
│   ├── Ocean of Tears.pdf
│   ├── Ocean of Tears.xml
│   ├── Ocean of Tears.mid
│   ├── Run Away.pdf
│   ├── Run Away.xml
│   ├── Run Away.mid
│   └── ...
└── songbook/
    └── Album Name - Complete Songbook.pdf

Upload everything from singles/ and songbook/ to your website. Don't upload source/ — those are working files with track number prefixes.

Website Page Structure

Album sheet music page should include:

• Brief description ("Free piano arrangements from [Album Name]")
• Per-track download links (PDF, XML, MIDI for each)
• Complete songbook download link
• Link to listen to the album
• Credit line: "Transcribed using AnthemScore, edited with MuseScore"

Hosting Options

Static site / existing website:

• Upload files to your web host
• Link directly to PDFs
• Simplest approach

GitHub Pages / Netlify:

• Free hosting for static files
• Version control built in
• Good for musician-developers

Bandcamp:

• Can include sheet music as bonus content with album purchase
• Or offer as free download

---

Preparing Your Songbook

Use the create_songbook MCP tool

create_songbook(album_slug, title="Album Name - Complete Songbook", page_size="letter")

Output includes:

• Title page (with album art)
• Copyright page
• Table of contents (track names + page numbers)
• All tracks in order

Front Matter Checklist

Title page (auto-generated):

• Book title
• Artist/composer name
• Subtitle (e.g., "Piano Arrangements")
• Album artwork (if provided)
• Website URL
• Year

Copyright page (auto-generated):

• Copyright notice
• Free distribution statement
• Transcription credits
• Website/contact info

Table of Contents (auto-generated):

• Track names with page numbers
• Leader dots for readability

Page Size

Most common: Letter (8.5" x 11") for home printing

---

Legal & Copyright

Copyright Ownership

If you wrote the songs:

• You own the copyright to music and lyrics
• Sheet music is a derivative work you also own
• You can distribute freely

If you sampled or covered songs:

• You do NOT own copyright
• Cannot distribute without permission
• Not applicable for original music

Copyright Notice

Included in songbook automatically:

© 2026 Your Artist Name.

This songbook is a free companion to the album.
Share freely. Credit appreciated.

Piano arrangements transcribed using AnthemScore.
Notation edited with MuseScore.

Creative Commons (Optional)

Consider adding a Creative Commons license for clear sharing terms:

• CC BY — anyone can use/share with credit
• CC BY-NC — non-commercial use with credit
• CC BY-SA — share-alike with credit

---

Alternative Platforms

Sheet Music Plus

Website: sheetmusicplus.com

• Digital marketplace for sheet music
• Sellers upload PDFs (individual tracks or books)
• Revenue split: ~70% to creator
• Best for: reaching sheet music enthusiasts

Musicnotes

Website: musicnotes.com

• Similar to Sheet Music Plus
• May require sample work for approval
• Digital marketplace with instant downloads

Print-on-Demand (Optional)

If you want physical copies available:

• Amazon KDP — print books, no upfront cost
• Lulu — print-on-demand
• Blurb — premium quality

These are optional — the primary distribution method is free digital download.

---

Licensing Considerations

Sync Licensing

Sheet music adds value to licensing packages:

• Shows professionalism
• Useful for music supervisors (TV/film)
• Helps arrangers/cover artists

Educational Use

Educators may use your sheet music for teaching:

• Make sure notation is accurate
• Include difficulty level in description
• Free distribution makes this more accessible

---

Checklist: Publishing Sheet Music

Before distributing:

• [ ] Singles prepared (clean titles, all 3 formats)
• [ ] Songbook PDF finalized (optional)
• [ ] Copyright notice included
• [ ] Album art on title page
• [ ] Track titles clean (no number prefixes)
• [ ] Review PDFs for readability

Website setup:

• [ ] Upload singles (PDF, XML, MIDI per track)
• [ ] Upload songbook PDF
• [ ] Create download page with track listing
• [ ] Add link to album streaming page
• [ ] Test all download links

Promotion:

• [ ] Announce on social media
• [ ] Add link to album descriptions
• [ ] Mention in YouTube video descriptions
• [ ] Post sample page images

---

Summary

Sheet music distribution workflow:

1. Generate sheet music with transcribe_audio (outputs to source/) 2. Prepare singles with prepare_singles (outputs to singles/) 3. Create songbook with create_songbook (outputs to songbook/) 4. Review all PDFs for quality 5. Upload to your website 6. Announce and promote

Benefits:

• Serves fans and musicians
• Adds value to licensing packages
• Professional presence
• Free distribution builds goodwill

Sheet Music Publisher - Software Requirements

This document lists all external software and dependencies needed for sheet music generation.

Required Software

1. AnthemScore (Audio Transcription)

Purpose: Converts audio files (WAV) to sheet music notation

Cost: $31-107 (one-time purchase, no subscription)

• Lite: $31 - Basic transcription, no editing
• Professional: $42 - Full editing + CLI ← Recommended
• Studio: $107 - Lifetime updates

Free Trial:

• 30 seconds per song
• 100 total transcriptions
• All features available

Download: https://www.lunaverus.com/

Platforms:

• macOS (10.13+)
• Windows (7, 8, 10, 11)
• Linux (Ubuntu, Fedora, Arch)

Installation Locations:

macOS:    /Applications/AnthemScore.app/Contents/MacOS/AnthemScore
Linux:    /usr/bin/anthemscore or /usr/local/bin/anthemscore
Windows:  C:\Program Files\AnthemScore\AnthemScore.exe

Why Professional Edition?

• CLI access (required for batch processing)
• Spectrogram editing
• Better accuracy controls
• Only $11 more than Lite

---

2. MuseScore (Notation Editing)

Purpose: Edit transcriptions, fix errors, export final PDFs

Cost: FREE (open source)

Download: https://musescore.org/

Platforms:

• macOS (10.12+)
• Windows (7, 8, 10, 11)
• Linux (AppImage, Ubuntu/Debian, Fedora, Arch)

Installation Locations:

macOS:    /Applications/MuseScore 4.app/Contents/MacOS/mscore
          /Applications/MuseScore 3.app/Contents/MacOS/mscore
Linux:    /usr/bin/musescore or /usr/bin/mscore
Windows:  C:\Program Files\MuseScore 4\bin\MuseScore4.exe

Current Versions:

• MuseScore 4 (latest, recommended)
• MuseScore 3 (stable, still supported)

Why Required?

• Re-export PDFs after title cleanup
• Manual polishing (fix wrong notes, add dynamics)
• Professional formatting

---

Python Dependencies

For songbook creation only (optional feature):

pip install pypdf reportlab pyyaml

Breakdown:

• pypdf - PDF manipulation (combine tracks into songbook)
• reportlab - Generate title page, copyright page, TOC
• pyyaml - Read config file for auto-detection

Not needed if you only want individual track PDFs.

---

Installation Guide

Quick Setup (macOS)

# 1. Install AnthemScore
# Download from https://www.lunaverus.com/ and install

# 2. Install MuseScore (via Homebrew)
brew install --cask musescore

# 3. Install Python dependencies
pip install pypdf reportlab pyyaml

# 4. Verify installation
ls -l "/Applications/AnthemScore.app/Contents/MacOS/AnthemScore"
ls -l "/Applications/MuseScore 4.app/Contents/MacOS/mscore"
python3 -c "import pypdf, reportlab, yaml; print('✓ All dependencies installed')"

Quick Setup (Linux - Ubuntu/Debian)

# 1. Install AnthemScore
# Download from https://www.lunaverus.com/ and install

# 2. Install MuseScore
sudo apt update
sudo apt install musescore

# 3. Install Python dependencies
pip install pypdf reportlab pyyaml

# 4. Verify installation
which anthemscore
which musescore
python3 -c "import pypdf, reportlab, yaml; print('✓ All dependencies installed')"

Quick Setup (Windows)

# 1. Install AnthemScore
# Download from https://www.lunaverus.com/ and install

# 2. Install MuseScore
# Download from https://musescore.org/ and install

# 3. Install Python dependencies
pip install pypdf reportlab pyyaml

# 4. Verify installation (PowerShell)
Test-Path "C:\Program Files\AnthemScore\AnthemScore.exe"
Test-Path "C:\Program Files\MuseScore 4\bin\MuseScore4.exe"
python -c "import pypdf, reportlab, yaml; print('✓ All dependencies installed')"

---

What Each Tool Does

Workflow Overview

1. WAV file
   ↓
2. AnthemScore (automated transcription)
   ↓
3. PDF + MusicXML generated
   ↓
4. MuseScore (optional manual polish)
   ↓
5. Title cleanup (MuseScore CLI re-exports PDFs)
   ↓
6. create_songbook.py (Python: combines PDFs)
   ↓
7. Final songbook PDF (KDP-ready)

Tool Responsibilities

---

Cost Summary

Alternatives:

• Use AnthemScore free trial (limited to 30 sec/song, 100 songs total)
• Skip songbook creation (saves Python dependencies)

---

Troubleshooting

"AnthemScore not found"

Check installation:

# macOS
ls -l "/Applications/AnthemScore.app/Contents/MacOS/AnthemScore"

# Linux
which anthemscore

# Windows
where anthemscore

If not found: 1. Verify AnthemScore is installed 2. Check installation location matches expected path 3. Add to PATH if installed to custom location

"MuseScore not found"

Check installation:

# macOS
ls -l "/Applications/MuseScore 4.app/Contents/MacOS/mscore"

# Linux
which musescore

# Windows
where mscore

If not found: 1. Verify MuseScore is installed 2. Check version (4 or 3) 3. Add to PATH if needed

"Python dependencies missing"

# Install
pip install pypdf reportlab pyyaml

# Verify
python3 -c "import pypdf; import reportlab; import yaml; print('OK')"

---

Optional: Add Custom Paths to Config

If you installed software to non-standard locations, you can add custom paths to ~/.bitwize-music/config.yaml:

# Future enhancement (not yet implemented)
tools:
  anthemscore: "/custom/path/to/AnthemScore"
  musescore: "/custom/path/to/mscore"

Current behavior: Scripts auto-detect standard install locations. If you have a non-standard install, add the executable to your PATH.

---

Summary

Minimum setup for sheet music generation: 1. Purchase AnthemScore Professional ($42) 2. Install MuseScore (free) 3. Install Python dependencies (free): pip install pypdf reportlab pyyaml

Total cost: $42 one-time

Free trial option: Use AnthemScore free trial for testing (30 sec/song limit)

Ready to use: Once installed, the /bitwize-music:sheet-music-publisher skill auto-detects everything.

Sheet Music Publisher - Workflow Detail

Detailed workflow phases, error handling, tips, and tool reference for the sheet-music-publisher skill.

---

Phase 1: Setup Verification

Check AnthemScore

macOS locations:

# Check standard install
ls -l "/Applications/AnthemScore.app/Contents/MacOS/AnthemScore"

# Try Homebrew
which anthemscore

Linux locations:

which anthemscore
ls -l /usr/bin/anthemscore /usr/local/bin/anthemscore

Windows locations:

C:\Program Files\AnthemScore\AnthemScore.exe
C:\Program Files (x86)\AnthemScore\AnthemScore.exe

If not found, tell user:

AnthemScore not found. Install from: https://www.lunaverus.com/

Recommended edition: Professional ($42) - full editing + CLI
Free trial available: 30 seconds per song, 100 total transcriptions

After installing, run this command again.

Check MuseScore

macOS locations:

ls -l "/Applications/MuseScore 4.app/Contents/MacOS/mscore"
ls -l "/Applications/MuseScore 3.app/Contents/MacOS/mscore"

Linux locations:

which mscore
ls -l /usr/bin/musescore /usr/local/bin/musescore

Windows locations:

C:\Program Files\MuseScore 4\bin\MuseScore4.exe
C:\Program Files\MuseScore 3\bin\MuseScore3.exe

If not found, tell user:

MuseScore not found. Install from: https://musescore.org/

Free and open source. Required for:
- Polishing transcriptions
- Fixing titles (re-exporting PDFs)
- Manual editing

After installing, run this command again.

Check Python Dependencies (Songbook Only)

If user wants to create a songbook, call get_python_command() to verify the venv exists, then check songbook deps:

{python_from_get_python_command} -c "import pypdf, reportlab"

If missing:

Songbook dependencies missing. Install with:

  ~/.bitwize-music/venv/bin/pip install pypdf reportlab

These are only needed for songbook creation (optional).

Phase 2: Track Selection

List mastered tracks:

ls -1 {audio_root}/artists/{artist}/albums/{genre}/{album}/*.wav

Ask user which tracks to transcribe:

Found N mastered WAV files.

Sheet music works best for:
✓ Melodic tracks (clear vocal/instrumental lines)
✓ Singer-songwriter, folk, acoustic
✓ Simple arrangements

Challenging:
⚠ Dense electronic music
⚠ Heavy distortion
⚠ Rapid rap vocals

Which tracks should I transcribe?
  1. All tracks
  2. Specific tracks (you'll select)
  3. Let me recommend based on metadata

If user chooses option 3, read track files from content directory to check for:

• Explicit vocal mentions in Suno Style Box
• Melodic genres (folk, country, singer-songwriter)
• Simple instrumentation descriptions

Recommend tracks that mention:

• "vocals", "singing", "melodic"
• "acoustic", "piano", "folk"
• Avoid: "aggressive", "distorted", "dense"

Phase 3: Automated Transcription

Run transcription via MCP:

transcribe_audio(album_slug)

The script will: 1. Read config to locate audio files 2. Find AnthemScore based on OS 3. Create output directory: {audio_root}/artists/{artist}/albums/{genre}/{album}/sheet-music/source/ 4. Process each WAV (~30-60 seconds per track) 5. Generate PDF + MusicXML + MIDI for each track

Monitor progress and report to user:

Transcribing tracks...

✓ Track 01: ocean-of-tears.wav - Complete
  → sheet-music/source/01-ocean-of-tears.pdf
  → sheet-music/source/01-ocean-of-tears.xml
  → sheet-music/source/01-ocean-of-tears.mid

✓ Track 02: run-away.wav - Complete
  → sheet-music/source/02-run-away.pdf
  → sheet-music/source/02-run-away.xml
  → sheet-music/source/02-run-away.mid

Transcription complete: N tracks processed

Phase 4: Quality Review & Polish

After transcription completes, recommend review:

Transcription complete. Generated PDFs are in:
  {audio_root}/artists/{artist}/albums/{genre}/{album}/sheet-music/

Next: Review transcriptions for accuracy

Automated transcription accuracy: 70-95% depending on arrangement

Recommend polish in MuseScore?
  - Fix notation errors (wrong notes, durations)
  - Add dynamics (p, mf, f)
  - Adjust key/time signatures
  - Format for print

Options:
  1. Yes - I'll polish in MuseScore now (pause here)
  2. Skip - PDFs are good enough
  3. Only specific tracks need polish (which ones?)

If user chooses polish:

Opening MusicXML files in MuseScore for editing...

Polish checklist:
  [ ] Key signature correct
  [ ] Time signature correct
  [ ] Note durations correct
  [ ] Rests complete measures
  [ ] Dynamics added (p, mf, f)
  [ ] Tempo marking added
  [ ] Title/composer in score properties

After editing in MuseScore:
  1. File → Export → PDF (300 DPI)
  2. Save MuseScore project (.mscz)
  3. Tell me when done: "polish complete"

I'll wait for your confirmation before proceeding.

Phase 5: Prepare Singles

After polish (or if skipped), prepare consumer-ready singles via MCP:

prepare_singles(album_slug)

What this does:

• Reads numbered source files from sheet-music/source/
• Applies smart title casing (e.g., "01-ocean-of-tears" → "Ocean of Tears")
• Updates MusicXML <work-title> tags
• Writes clean-titled files to sheet-music/singles/ (PDF, XML, MIDI)
• Creates .manifest.json for track ordering
• Source files are never modified

Report:

Preparing singles...

✓ Prepared 5 tracks in singles/
✓ Created .manifest.json

Files ready for website distribution.

Phase 6: Songbook Creation (Optional)

Ask user:

Create a combined songbook PDF?

This creates a distribution-ready book with:
  - Title page (with album art if available)
  - Copyright page
  - Table of contents
  - All tracks in order

Perfect for:
  - Website distribution (free download)
  - Licensing packages
  - Fan resources

Create songbook? [Y/n]

If yes, run via MCP:

create_songbook(album_slug, title="{album_title} Songbook", page_size="letter")

Detect page size from config:

• Read ~/.bitwize-music/config.yaml → sheet_music.page_size
• Default: letter

Detect cover art:

# Check standard location
ls -l {audio_root}/artists/{artist}/albums/{genre}/{album}/album.png

Report:

Creating songbook...

✓ Found album art: album.png
✓ Songbook created: {album}/sheet-music/{Album}_Songbook.pdf

Contents:
  - Title page (with album art)
  - Copyright page
  - Table of contents (N tracks)
  - N tracks (XX pages total)

Ready for website distribution.

Phase 7: Summary & Next Steps

Final summary:

✅ SHEET MUSIC GENERATION COMPLETE

Album: {album_name}
Tracks transcribed: N
Output: {audio_root}/artists/{artist}/albums/{genre}/{album}/sheet-music/

Files generated:
  - N × PDF (individual tracks)
  - N × MusicXML (editable source)
  [- 1 × Songbook PDF (combined)]

Next steps:
  1. Review PDFs for accuracy
  2. Upload singles + songbook to your website
  3. [Optional] Include in licensing packages
  4. [Optional] Share on social media

See publishing guide for distribution instructions:
  ${CLAUDE_PLUGIN_ROOT}/skills/sheet-music-publisher/publishing-guide.md

Error Handling

AnthemScore Not Found

Error message:

AnthemScore not detected on your system.

Install from: https://www.lunaverus.com/

Editions:
  - Lite: $31 (basic transcription, no editing)
  - Professional: $42 (full editing + CLI) ← Recommended
  - Studio: $107 (lifetime updates)

Free trial: 30 seconds per song, 100 total transcriptions

After installing, run this command again.

MuseScore Not Found

If only needed for title fixing:

MuseScore not found. Options:

1. Install MuseScore (free): https://musescore.org/
   Then I'll fix titles and re-export PDFs

2. Skip title fixing
   PDFs will have track numbers in titles (e.g., "01 - Song Name")
   You can fix this manually later

Which option?

No WAV Files Found

No WAV files found in: {audio_root}/artists/{artist}/albums/{genre}/{album}/

Expected location based on config:
  audio_root: {audio_root}
  artist: {artist}
  album: {album}

Check:
  1. Album name spelled correctly?
  2. Audio files in correct location?
  3. Files are .wav format (not .mp3)?

Use /bitwize-music:import-audio to move files if needed.

Python Dependencies Missing

Songbook dependencies missing. Install with:

  pip install pypdf reportlab

Note: Only needed for songbook creation. Individual PDFs are already generated.

Install and retry? [Y/n]

Track Transcription Failed

If AnthemScore returns error:

⚠ Track failed: {track_name}

Possible causes:
  - File is corrupted or unreadable
  - Format not supported (use WAV)
  - AnthemScore license issue

Try:
  1. Verify file plays in audio player
  2. Re-export from DAW as WAV
  3. Check AnthemScore license status

Skip this track and continue? [Y/n]

Tips for Better Results

Audio Quality Matters

Tell user:

Better source = better transcription

Best practices:
  ✓ Use mastered WAV files (not MP3)
  ✓ Ensure no clipping
  ✓ Clear, well-mixed audio
  ✗ Avoid overly compressed audio
  ✗ Avoid heavy distortion

When to Recommend Stem Separation

If user has dense/complex tracks, suggest:

For complex arrangements, stem separation can improve accuracy:

1. Install Demucs: pip install demucs
2. Separate stems: demucs {track.wav}
3. Transcribe vocals.wav and other.wav separately
4. Combine in MuseScore

Use for:
  - Dense rock/pop arrangements
  - Buried melodies
  - Professional licensing needs

Skip for:
  - Simple acoustic tracks
  - Quick reference sheets

Want to try stem separation? [y/N]

Piano Reduction Philosophy

If user asks about transcription approach:

AnthemScore creates "piano reduction" - a playable piano arrangement.

It captures:
  ✓ Main melody
  ✓ Bass line
  ✓ Harmonic structure
  ✓ Characteristic riffs

It simplifies/omits:
  ✗ Drum patterns
  ✗ Doubled/layered parts
  ✗ Complex backgrounds
  ✗ Sound effects

Goal: Playable by intermediate pianist, captures song essence

Integration with Main Workflow

When to proactively offer this skill:

1. After album mastering complete:

• User says "mastering done"
• All tracks marked Final

→ Ask: "Want to generate sheet music for this album?"

2. User mentions licensing:

• "I want to license this"
• "preparing licensing package"

→ Suggest: "Sheet music is valuable for sync licensing. Generate now?"

3. User mentions publishing or distribution:

• "distribute sheet music"
• "create a songbook"

→ Invoke this skill automatically

Position in workflow:

Generate → Master → [Sheet Music] → Release
                        ↑
                   Optional enhancement

Config Integration

Always read config first:

cat ~/.bitwize-music/config.yaml

Extract values:

• artist.name → For songbook metadata
• paths.audio_root → Where mastered audio lives
• paths.content_root → Where album metadata is
• urls.soundcloud (or other) → For website field in songbook
• sheet_music.page_size → For songbook dimensions (default: letter)
• sheet_music.section_headers → Whether to add section headers (default: false)

Path construction:

Input audio:  {audio_root}/artists/{artist}/albums/{genre}/{album}/*.wav
Output:       {audio_root}/artists/{artist}/albums/{genre}/{album}/sheet-music/

MCP Tool Reference

transcribe_audio

# All tracks
transcribe_audio(album_slug)

# Single track
transcribe_audio(album_slug, track_filename="01-track-name.wav")

# PDF only (no MusicXML)
transcribe_audio(album_slug, formats="pdf")

# Dry run
transcribe_audio(album_slug, dry_run=True)

prepare_singles

# Prepare clean-titled singles (PDF, XML, MIDI)
prepare_singles(album_slug)

# Dry run (preview only)
prepare_singles(album_slug, dry_run=True)

# XML only (skip PDF/MIDI)
prepare_singles(album_slug, xml_only=True)

create_songbook

# Standard songbook
create_songbook(album_slug, title="Sample Album Songbook")

# With page size
create_songbook(album_slug, title="Sample Album Songbook", page_size="9x12")

Quality Standards

Publishing-ready sheet music includes:

• ✓ Clean, readable notation
• ✓ Correct key/time signatures
• ✓ Proper credits and copyright
• ✓ Playable by intermediate pianist
• ✓ No track number prefixes in titles
• ✓ Consistent formatting

Minimum viable:

• ✓ Correct notes (mostly)
• ✓ Readable layout
• ✓ Title and composer
• ~ May need polish for errors

Set user expectations:

Automated transcription accuracy: 70-95%

Perfect for:
  - Reference sheets
  - Licensing packages
  - Fan resources

May need polish for:
  - Professional publishing
  - Teaching materials
  - Critical accuracy needs

Workflow State Tracking

Update album README after sheet music generation:

If user confirms "done with sheet music", suggest:

Update Album Completion Checklist?

Add:
  - [✓] Sheet music generated (N tracks)

This helps track release readiness.