« All Events

Paper Session 2a: Music Information Retrieval

May 11 @ 11:00 am - 12:30 pm
Three papers will be presented and discussed:

 

Axel Berndt, Aida Amiryan-Stein, Manuel Peters, Meinard Müller and Stefan Balke, “ChoraleWind: An Expressive Wind-Quartet Dataset for End-to-End Rendering from the Neues Thüringer Choralbuch”
We introduce ChoraleWind, a dataset along with a framework for a reproducible end-to-end rendering from the Neues Thüringer Choralbuch (NTCB). The dataset comprises 311 four-part chorales and covers the full pipeline from symbolic score encoding to performance-level rendition and synthesized audio. ChoraleWind includes a rule-based performance model that generates expressive timing, dynamics, and articulation, including metric and structural accents as well as phrase-end gestures from high-quality MEI encoding of the NTCB chorales, combined with a wind-instrument synthesis based on physical modeling that produces isolated stems and ensemble mixes. The dataset provides aligned symbolic representations, performance annotations, and multitrack audio, enabling systematic training and evaluation of score-to-audio wind-quartet rendering methods under fully controlled conditions. Rather than aiming at state-of-the-art purely data-driven synthesis, ChoraleWind is designed as a transparent and reproducible testbed for studying expressive performance generation, timbre modeling, and evaluation of wind-quartet rendering systems.
Mário Pereira, António Sá Pinto, Treasa Harkin and Gilberto Bernardes, “Computational Analysis of Expressive Tempo in Irish Traditional Dance Music”
This paper presents a computational study of expressive tempo in Irish traditional dance music, analysing 136 annotated performances of reels and jigs. Using beat-level tempo calculation, predominant-tempo estimation, and deviation-curve analysis, we examine how timing varies across tune types, performance settings, and musical structure. Results show that expressive deviations are generally subtle: reels display a mild deceleration tendency, jigs remain highly tempo-stable,
and solo–ensemble and instrument-specific differences are minimal. Phrase-level clustering reveals three characteristic deviation profiles, with strong acceleration occurring only in opening phrases, reflecting common slow-start performance practices. These findings provide, to the best of our knowledge, the first systematic quantitative characterisation of expressive timing in this tradition and highlight how micro-variations emerge from stylistic, technical, and interpretive factors while maintaining overall temporal stability.
Gilberto Bernardes, Nádia Moura and António Sá Pinto, “Perpetual Dialogues: A Computational Analysis of Voice–Guitar Interaction in Carlos Paredes’s Discography”

Computational musicology enables systematic analysis of performative and structural traits in recorded music, yet existing approaches remain largely tailored to notated, score-based repertoires. This study advances a methodology for analyzing voice–guitar interaction in Carlos Paredes’s vocal collaborations—an oral-tradition context where compositional and performative layers co-emerge.
Using source-separated stems, physics-informed harmonic modeling, and beat-level audio descriptors, we examine melodic, harmonic, and rhythmic relationships across eight recordings with four singers. Our commonality–diversity framework, combining multi-scale correlation analysis with residual-based detection of structural deviations, reveals that expressive coordination is predominantly piece-specific rather than corpus-wide. Diversity events systematically align with formal boundaries and textural shifts, demonstrating that the proposed approach can identify musically salient reorganizations with minimal human annotation. The framework further offers a generalizable computational strategy for repertoires without notated blueprints, extending Music Performance Analysis into oral-tradition and improvisation-inflected practices.

 

Details

Organizer

Venue