TART: A Modular Tool for Technique-Aware Audio-to-Tablature Guitar Transcription
This is a new academic paper that introduces TART, a framework for transcribing guitar audio into tablature with expressive techniques. It's the first to generate tablature with fingering and technique annotations directly from audio.
TART is a new framework for transcribing guitar audio into tablature. It addresses three key challenges: capturing expressive techniques like slides and bends, assigning notes to the correct string-fret combinations, and working with noisy real-world audio. The system is a four-stage pipeline that includes an audio-to-MIDI model, a technique classifier, a string-fret assignment model, and a tablature generator. It was evaluated on GuitarSet, EGDB, and two augmented benchmarks, achieving 81.35% audio-to-MIDI F50 and 71.8% string-fret Tab F1, outperforming prior baselines.
Automatic Music Transcription (AMT) for guitar remains limited by three challenges: existing systems often fail to capture expressive techniques such as slides, bends, and percussive hits; they often assign notes to incorrect string-fret combinations; and they are typically trained on clean recordings, limiting their generalization to noisy real-world audio. To address these challenges, we propose TART, a modular four-stage audio-to-tablature pipeline consisting of (1) an audio-to-MIDI transcription model, (2) an expressive technique classifier, (3) an audio-conditioned T5 encoder-decoder for string-fret assignment, and (4) an automated tablature generator. We evaluate TART in a zero-shot setting on GuitarSet, EGDB, and two augmented benchmarks, Noisy GuitarSet and Noisy EGDB. Averaged across these four benchmarks, TART achieves 81.35% audio-to-MIDI F50 (+6.67 points over the best prior baseline), 71.8% string-fret Tab F1 (+8.5 points over the best prior baseline), and 54.08% end-to-end Tab F1. To our knowledge, TART is the first framework to generate guitar tablature with both fingering and expressive technique annotations directly from guitar audio.
