Classical Guitar Duet Separation using GuitarDuets -- a Dataset of Real and Synthesized Guitar Recordings

TL;DR

This paper introduces GuitarDuets, a new dataset of real and synthesized classical guitar duets, and adapts state-of-the-art separation models to improve monotimbral guitar duet source separation, highlighting the benefits of combined data and note information.

Contribution

The work provides the GuitarDuets dataset, adapts Demucs for monotimbral MSS, and develops a joint transcription and separation framework for classical guitar duets.

Findings

Using both real and synthesized data improves separation performance.

Ground-truth note labels significantly enhance separation results.

Predicted note estimates offer only marginal improvements.

Abstract

Recent advancements in music source separation (MSS) have focused in the multi-timbral case, with existing architectures tailored for the separation of distinct instruments, overlooking thus the challenge of separating instruments with similar timbral characteristics. Addressing this gap, our work focuses on monotimbral MSS, specifically within the context of classical guitar duets. To this end, we introduce the GuitarDuets dataset, featuring a combined total of approximately three hours of real and synthesized classical guitar duet recordings, as well as note-level annotations of the synthesized duets. We perform an extensive cross-dataset evaluation by adapting Demucs, a state-of-the-art MSS architecture, to monotimbral source separation. Furthermore, we develop a joint permutation-invariant transcription and separation framework, to exploit note event predictions as auxiliary information. Our results indicate that utilizing both the real and synthesized subsets of GuitarDuets leads to improved separation performance in an independently recorded test set compared to utilizing solely one subset. We also find that while the availability of ground-truth note labels greatly helps the performance of the separation network, the predicted note estimates result only in marginal improvement. Finally, we discuss the behavior of commonly utilized metrics, such as SDR and SI-SDR, in the context of monotimbral MSS.