Pitch Spelling Jazz Lead Sheets, Solo Transcriptions, Classical Piano and Monophonic Scores

2026-06-18 · Source: Computation and Language · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Music Technology · Depth: Advanced, quick

Summary

An algorithm has been developed for automated pitch spelling and key estimation, processing MIDI-like input that includes note pitches in semitones and bar boundaries. This system jointly evaluates related information elements through a two-stage optimization process. The initial "modal" stage proposes a probable scale for each bar using a shortest-path search, aiming to minimize accidentals in the printed score. Subsequently, the "tonal" stage utilizes these local scales to determine the global Key Signature and note names, optimizing musical notation for the entire piece. Evaluations were performed on diverse digital musical scores, including jazz lead sheets from the Real Book, jazz solo and bass line transcriptions, traditional tunes, and classical scores for piano and monophonic instruments. This method was primarily designed for music transcription, particularly for building digital jazz solo collections from audio recordings, supporting music analysis, teaching, and cultural heritage preservation, and also introduces new distances between jazz scales for musicological research.

Key takeaway

For Creative Technologists or Research Scientists developing music transcription tools, this algorithm offers a robust approach to automate pitch spelling and key estimation. You can leverage its two-stage optimization to produce musically accurate notation from MIDI-like inputs, significantly streamlining the creation of digital archives for jazz solos or classical pieces. Consider integrating this method to enhance the fidelity of your automated transcription systems and support musicological studies.

Key insights

The algorithm optimizes pitch spelling and key estimation using a two-stage modal and tonal process for diverse musical scores.

Principles

• Jointly evaluate pitch spelling and key estimation.
• Minimize accidentals via shortest-path search.
• Local scales inform global key signature.

Method

The algorithm uses a "modal" stage for local scale proposal via shortest-path search, followed by a "tonal" stage to estimate global Key Signature and note names.

In practice

• Transcribe jazz solos from audio recordings.
• Build digital music analysis collections.
• Define new jazz scale distances.

Topics

• Pitch Spelling
• Key Estimation
• Music Transcription
• Jazz Lead Sheets
• Classical Scores
• Music Analysis

Best for: AI Scientist, Research Scientist, Creative Technologist

Related on AIssential

• Tonal parsimony in chord-sequence analysis: combining modulation cost and tonal vocabulary — Tonal parsimony optimizes chord sequence analysis by minimizing modulations and distinct tonal centers.
• Beyond Artifacts: Towards Generalizable Synthetic Song Detection via Music-Intrinsic Features — Sofia improves synthetic song detection by modeling music-intrinsic features with an adaptive Mixture-of-Experts framework.
• Hybrid Diffusion Transformer for Instruction-Guided Audio Editing via Rectified Flow — The Hybrid Diffusion Transformer uses a two-stage coarse-to-fine strategy for efficient, accurate instruction-guided audio editing.
• Automatic Speech Recognition for Child Reading: A Phonemic Approach using Isolated Words in Brazilian Portuguese — A phonemic approach with fine-tuned Wav2Vec2.0 significantly improves ASR for child reading assessment in Brazilian Portuguese.
• LoSATok: Low-dimensional Semantic-Acoustic Tokenizer for Cross-Domain Audio Understanding and Generation — LoSATok unifies audio understanding and generation via low-dimensional semantic-acoustic tokens, reducing Diffusion Transformer modeling complexity.
• A Dataset for the Recognition of Historical and Handwritten Music Scores in Western Notation — MusiCorpus provides the largest dataset of historical handwritten music for Optical Music Recognition.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Computation and Language.