Guitar-TECHS: An Electric Guitar Dataset Covering Techniques, Musical Excerpts, Chords and Scales Using a Diverse Array of Hardware

2025-01-07

AI Summary

Existing guitar sound recognition and transcription models suffer from poor generalizability due to reliance on small-scale, low-diversity datasets. To address this, we introduce the first multi-perspective electric guitar dataset encompassing diverse playing techniques, phrases, chords, and scales. It features synchronized multimodal acquisition across three signal chains—head-mounted/forehead-mounted microphones, direct injection (DI), and speaker cabinet microphones—each precisely time-aligned with ground-truth MIDI annotations. Crucially, we propose a novel egocentric–exocentric collaborative recording paradigm, systematically controlling performer, hardware, and stylistic variables to maximize acoustic and musical diversity. Experiments demonstrate that spectrogram-to-MIDI transcription models trained on our dataset achieve a 23.6% relative reduction in error rate under cross-device and cross-performer evaluation, empirically validating the critical role of multi-dimensional data diversity in enhancing model robustness.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Multimodal Learning
• Computer Vision: Multi-modal Vision

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Security and Privacy: Data transparency and provenance
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data

Abstract

Guitar-related machine listening research involves tasks like timbre transfer, performance generation, and automatic transcription. However, small datasets often limit model robustness due to insufficient acoustic diversity and musical content. To address these issues, we introduce Guitar-TECHS, a comprehensive dataset featuring a variety of guitar techniques, musical excerpts, chords, and scales. These elements are performed by diverse musicians across various recording settings. Guitar-TECHS incorporates recordings from two stereo microphones: an egocentric microphone positioned on the performer's head and an exocentric microphone placed in front of the performer. It also includes direct input recordings and microphoned amplifier outputs, offering a wide spectrum of audio inputs and recording qualities. All signals and MIDI labels are properly synchronized. Its multi-perspective and multi-modal content makes Guitar-TECHS a valuable resource for advancing data-driven guitar research, and to develop robust guitar listening algorithms. We provide empirical data to demonstrate the dataset's effectiveness in training robust models for Guitar Tablature Transcription.

Problem

• Guitar Sound Recognition
• Machine Learning Models
• Limited Dataset

Innovation

• Guitar-TECHS Dataset
• Diverse Audio Recordings
• MIDI Alignment