GAPS: A Large and Diverse Classical Guitar Dataset and Benchmark Transcription Model

2024/08/16 by Riley, Xavier, Guo, Zixun, Edwards, Drew +1 · 4 citations
#Audio and Speech Processing (eess.AS) #FOS: Computer and information sciences #FOS: Electrical engineering #Sound (cs.SD) #electronic engineering #information engineering

paper · doi:10.48550/arxiv.2408.08653

Abstract

We introduce GAPS (Guitar-Aligned Performance Scores), a new dataset of classical guitar performances, and a benchmark guitar transcription model that achieves state-of-the-art performance on GuitarSet in both supervised and zero-shot settings. GAPS is the largest dataset of real guitar audio, containing 14 hours of freely available audio-score aligned pairs, recorded in diverse conditions by over 200 performers, together with high-resolution note-level MIDI alignments and performance videos. These enable us to train a state-of-the-art model for automatic transcription of solo guitar recordings which can generalise well to real world audio that is unseen during training.

Cited by

• Count The Notes: Histogram-Based Supervision for Automatic Music Transcription
• Ethics Statements in AI Music Papers: The Effective and the Ineffective
• Score-Informed BiLSTM Correction for Refining MIDI Velocity in Automatic Piano Transcription
• Filling MIDI Velocity using U-Net Image Colorizer

Related

• GOAT: A Large Dataset of Paired Guitar Audio Recordings and Tablatures
• High Resolution Guitar Transcription via Domain Adaptation
• SCORE-SET: A dataset of GuitarPro files for Music Phrase Generation and Sequence Learning
• Classical Guitar Duet Separation using GuitarDuets -- a Dataset of Real and Synthesized Guitar Recordings
• SynthTab: Leveraging Synthesized Data for Guitar Tablature Transcription