Abstract
The development of deep learning algorithms has extensively empowered humanity's task automatization capacity. However, the huge improvement in the performance of these models is highly correlated with their increasing level of complexity, limiting their usefulness in human-oriented applications, which are usually deployed in resource-constrained devices. This led to the development of compression techniques that drastically reduce the computational and memory costs of deep learning models without significant performance degradation. These compressed models are especially essential when implementing multi-model fusion solutions where multiple models are required to operate simultaneously. This paper aims to systematize the current literature on this topic by presenting a comprehensive survey of model compression techniques in biometrics applications, namely quantization, knowledge distillation and pruning. We conduct a critical analysis of the comparative value of these techniques, focusing on their advantages and disadvantages and presenting suggestions for future work directions that can potentially improve the current methods. Additionally, we discuss and analyze the link between model bias and model compression, highlighting the need to direct compression research toward model fairness in future works.

Abstract
The rapid proliferation of Edge AI applications demands efficient, low-power computing architectures tailored to specific workloads. The RISC-V ecosystem is a promising solution, and has led to a fast growth of implementations based on custom instructions extensions, but with varying degrees of functionality and support which may hinder easy adoption. In this paper, we extensively review existing RISC-V extensions targeting primarily the AI domain and respective compilation flows, highlighting challenges in deployment, usability, and compatibility. We further implement and provide usable containerized environments for two of these works. To address the identified challenges, we then propose an approach for lightweight early validation of custom instructions via source-to-source transformations, without need of compiler modifications. We target our own Single Instruction Multiple Data (SIMD) accelerator, which we integrate into a CORE-V cv32e40px baseline core through custom instructions, and versus which we achieve up to 11.9x speedup for matrix-vector operations.

Abstract
Abstract In this article, we formalize and analyze qualia motion, i.e., the process by which a composition transitions across distinct harmonic qualities through the Fourier qualia space (FQS)—a multidimensional and transposition-independent space based on the discrete Fourier transform (DFT) coefficients’ magnitude. In the FQS, the plot of set classes relies on their harmonic qualities—such as diatonicity and octatonicity—enabling us to (1) identify the pitch-class set in a musical phrase that best represents its qualia—a reference sonority; (2) define a harmonic progression using all sequential reference sonorities in a piece; (3) visualize trajectory in space; and (4) establish a statistical metric for the ambiguity of harmonic qualia. Finally, we discuss Schoenberg's Op. 19, No. 1, analyzing the sense of its harmonic path. The proposed space leverages a bipartite, symmetrical, and consequential structure and unveils ambiguity as an element of nondirected linearity and counterpoint.

Abstract
This paper introduces Motiv, a dataset of expert saxophonist recordings illustrating parallel, similar, oblique, and contrary motions. These motions are variations of three phrases from Jesús Villa-Rojo's "Lamento,"with controlled similarities. The dataset includes 116 audio samples recorded by four tenor saxophonists, each annotated with descriptions of motions, musical scores, and latent space vectors generated using the VocalSet RAVE model. Motiv enables the analysis of motion types and their geometric relationships in latent spaces. Our preliminary dataset analysis shows that parallel motions align closely with original phrases, while contrary motions exhibit the largest deviations, and oblique motions show mixed patterns. The dataset also highlights the impact of individual performer nuances. Motiv supports a variety of music information retrieval (MIR) tasks, including gesture-based recognition, performance analysis, and motion-driven retrieval. It also provides insights into the relationship between human motion and music, contributing to real-time music interaction and automated performance systems. © 2025 Copyright held by the owner/author(s).

Abstract
State-of-the-art symbolic music generation models have recently achieved remarkable output quality, yet explicit control over compositional features, such as tonal tension, remains challenging. We propose a novel approach that integrates a computational tonal tension model, based on tonal interval vector analysis, into a Transformer framework. Our method employs a two-level beam search strategy during inference. At the token level, generated candidates are re-ranked using model probability and diversity metrics to maintain overall quality. At the bar level, a tension-based re-ranking is applied to ensure that the generated music aligns with a desired tension curve. Objective evaluations indicate that our approach effectively modulates tonal tension, and subjective listening tests confirm that the system produces outputs that align with the target tension. These results demonstrate that explicit tension conditioning through a dual-level beam search provides a powerful and intuitive tool to guide AI-generated music. Furthermore, our experiments demonstrate that our method can generate multiple distinct musical interpretations under the same tension condition.

Abstract
We present a metadata enrichment framework for Music Encoding Initiative (MEI) files, featuring mid- to higher-level multimodal features to support content-driven (similarity) retrieval with semantic awareness across large collections. While traditional metadata captures basic bibliographic and structural elements, it often lacks the depth required for advanced retrieval tasks that rely on musical phrases, form, key or mode, idiosyncratic patterns, and textual topics. To address this, we propose a system that fosters the computational analysis and edition of MEI encodings at scale. Inserting extended metadata derived from computational analysis and heuristic rules lays the groundwork for more nuanced retrieval tools. A batch environment and a lightweight JavaScript web-based application propose a complementary workflow by offering large-scale annotations and an interactive environment for reviewing, validating, and refining MEI files' metadata. Development is informed by user-centered methodologies, including consultations with music editors and digital musicologists, and has been co-designed in the context of orally transmitted folk music traditions, ensuring that both the batch processes and interactive tools align with scholarly and domain-specific needs.