1,720,987 research outputs found
Comparing affective and sensorial expressions in music performance by machine learning techniques
Full text linkScore-Independent Audio Features for Description of Music Expression
No full textDuring a music performance, the musician adds expressiveness to the musical message by changing timing, dynamics, and timbre of the musical events to communicate an expressive intention. Traditionally, the analysis of music expression is based on measurements of the deviations of the acoustic parameters with respect to the written score. In this paper, we employ machine learning techniques to understand the expressive communication and to derive audio features at an intermediate level, between music intended as a structured language and notes intended as sound at a more physical level. We start by extracting audio features from expressive performances that were recorded by asking the musicians to perform in order to convey different expressive intentions. We use a sequential forward selection procedure to rank and select a set of features for a general description of the expressions, and a second one specific for each instrument. We show that higher recognition ratings are achieved by using a set of four features which can be specifically related to qualitative descriptions of the sound by physical metaphors. These audio features can be used to retrieve expressive content on audio data, and to design the next generation of search engines for music information retrieval
Expressiveness detection of music performances in the Kinematics Energy Space
Full text linkAn experiment for the automatic detection of expressive- ness in music performances using a perceptive based audi- tory models is presented. We recognize the intentions with reference to the Kinematics Energy expressive space. Au- dio features have been firstly extracted using a perception- based analysis, then we have made several analyses on timing and spectral features over overlapping sliding win- dows, estimating average and variance for each one of the features. Using a naive Bayesian classifier we investigated which features are most relevant for expression detection. This experiment also yielded interesting contributions for tuning the Kinematics Energy space with new features
Computational models for audio expressive communication
Full text linkAudio objects have an important role in multimedia communication, and audio expressive content can enrich the Human Computer interaction in multimedia systems. In this paper expressive analysis, modelling and detection paradigms are reviewed and future research efforts are discussed
Personalized 3d sound rendering for content creation, delivery, and presentation.
Full text linkAdvanced models for 3D audio rendering are increasingly needed in the networked electronic media world, and play a central role within the strategic research objectives identified in the NEM research agenda. This paper presents a model for sound spatialization which includes additional features with respect to existing systems, being parametrized according to anthropometric information of the user, and being based on audio processing with low-order filters, thus allowing for significant reduction of the computational costs. This technology can offer a transversal contribution to the NEM research objectives, with respect to content creation and adaptation, intelligent delivery and augmented media presentation, by improving the quality of the immersive experience in a number of contexts where realistic spatialization and personalised sound reproduction is a key requirement, in particular in mobile contexts with headphone-based rendering
Toward an action based metaphor for gestural interaction with musical contents.
No full textTechnology-mediated music access is more and more becoming an interactive process, involving non-linguistic communication and action-based modalities. A better understanding of the musical experience and how this experience can be described is a crucial issue to render more effective and natural the interaction with musical contents. This paper aims at verifying if and how non-linguistic descriptors can be related to musical stimuli and their expressive cues. We designed four experiments using two sets of musical stimuli (simple and complex) and two sets of non-linguistic descriptors (acoustic and haptic), that we called attractors. In particular, the haptic attractors simulate the mechanic concepts of friction, elasticity and inertia (FEI). The results showed that subjects are able to relate musical stimuli with both acoustic and haptic attractors, even if the FEI metaphor seems to be more suitable for describing expressive cues in simple musical excerpts, where the expressive content is mainly related to performance cues, than in complex musical stimuli, where musical structure is more relevant
Exploring similarities of affective and sensorial expressive intentions in music performance
Full text linkPhysical and perceptual similarities of affective and
sensorial expressive intentions in music performance
are explored. Machine learning techniques were employed
to select and validate the most relevant low level
features and an interpretation of the clustered organization
based on action and physical analogy is proposed.
A perceptual experiment is then presented which confirms
the same groupings and suggests an intrinsic correspondence
of affective and sensorial expressive intentions
Perceptual organization of affective and sensorial expressive intentions in music performance.
No full textExpression communication is the added value of a musical performance. It is part of the reason why music is interesting to listen to and sounds alive. Previous work on the analysis of acoustical features yielded relevant features for the recognition of different expressive intentions, inspired both by emotional and sensorial adjectives. In this article, machine learning techniques are employed to understand how expressive performances represented by the selected features are clustered on a low-dimensional space, and to define a measure of acoustical similarity. Being that expressive intentions are similar according to the features used for the recognition, and since recognition implies subjective evaluation, we hypothesized that performances are similar also from a perceptual point of view. We then compared and integrated the clustering of acoustical features with the results of two listening experiments. A first experiment aims at verifying whether subjects can distinguish different categories of expressive intentions, and a second experiment aims at understanding which expressions are perceptually clustered together in order to derive common evaluation criteria adopted by listeners, and to obtain the perceptual organization of affective and sensorial expressive intentions. An interpretation of the resulting spatial representation based on action is proposed and discussed
A dynamic analogy between integro-differential operators and musical expressiveness
No full textMusic is often related to mathematics. Since Pythagoras, the focus is mainly on the relational and structural aspects of pitches described by arithmetic or geometric theories, and on the sound production and propagation described by differential equation, Fourier analysis and computer algorithms. However, music is not only score or sound; it conveys emotional and affective content. The aim of this paper is to explore a possible association between musical expressiveness and basic physical phenomena described by integro-differential operators
Music Expression Understanding Based on a Joint Semantic Space
No full textA paradigm for music expression understanding based on a joint semantic space, described by both affective and sensorial adjectives, is presented. Machine learning techniques were employed to select and validate relevant low level features, and an interpretation of the clustered organization based on action and physical analogy is proposed
- …
