Sergi Jorda - Academia.edu (original) (raw)

Papers by Sergi Jorda

Research paper thumbnail of SLFiducials

In this paper we introduce a new method for 6DoF marker tracking, specially designed for Microsof... more In this paper we introduce a new method for 6DoF marker tracking, specially designed for Microsoft SecondLight or any camera-based tabletop interface that is able to see objects through the surface. Our method is based on topological region adjacency for the identification of the markers, which are fitted into a squared shape for properly track the marker pose in the real world. We also describe the constraints imposed by the system which will determine the size and ID range of the new markers, and we finally evaluate the system.

Research paper thumbnail of Audio digital y MIDI

Research paper thumbnail of Target-based rhythmic pattern generation and variation with genetic algorithms

Composing drum patterns and musically developing them through repetition and variation is a typic... more Composing drum patterns and musically developing them through repetition and variation is a typical task in electronic music production. We propose a system that, given an input pattern, automatically creates related patterns using a genetic algorithm. Two distance measures (the Hamming distance and directed-swap distance) that relate to rhythmic similarity are shown to derive usable fitness functions for the algorithm. A software instrument in the Max for Live environment presents how this can be used in real musical applications. Finally, a user survey was carried out to examine and compare the effectiveness of the fitness metrics in determining rhythmic similarity as well as the usefulness of the instrument for musical creation.

Research paper thumbnail of Proceedings of the 7th International Conference on Tangible, Embedded and Embodied Interaction

Tangible and Embedded Interaction, Feb 10, 2013

A cohesive crack model is proposed to describe strain localization for the materials where strain... more A cohesive crack model is proposed to describe strain localization for the materials where strain-hardening is not prevailing over strain-softening (geomaterials, concrete-like materials, ceramics, etc.). Such a model is able to predict the size effects of fracture mechanics, i.e., the transition from ductile to brittle structure behaviour by increasing the size scale and keeping the geometrical shape unchanged. Whereas for Mode I, only untieing of the finite element nodes is applied to simulate crack growth, for Mixed Mode interelement crack propagation a topological variation is required at each step. In the case of four point shear testing, the load vs. deflection diagrams reveal snap-back instability for large sizes. By increasing the specimen sizes, such instability tends to reproduce the classical LEFM instability, predicted by the Maximum Circumferential Stress Criterion. Experimentally, the fracture toughness parameter of concrete appears to be unique and represented by the Mode I fracture energy Gp or the stress-intensity factor Kic even for Mixed Mode problems.

Research paper thumbnail of Prototyping interactions with Online Multimodal Repositories and Interactive Machine Learning

Interaction designers often use machine learning tools to generate intuitive mappings between com... more Interaction designers often use machine learning tools to generate intuitive mappings between complex inputs and outputs. These tools are usually trained live, which is not always feasible or practical. We combine RepoVizz, an online repository and visualizer for multimodal data, with a suite of Interactive Machine Learning tools, to demonstrate a technical solution for prototyping multimodal interactions that decouples the data acquisition step from the model training step. This way, different input data set-ups can be easily replicated, shared and experimented upon their capability to control complex output without the need to repeat the technical set-up.

Research paper thumbnail of Proceedings of the Sixth International Conference on Tangible, Embedded and Embodied Interaction

TEI is already in its seventh edition. This conference now counts with a solid and mature communi... more TEI is already in its seventh edition. This conference now counts with a solid and mature community, which is nonetheless growing every year. We are very happy to host this conference in Barcelona with one of the highest registrations ever, proving the high acceptance of the conference and its scientific value. We are also very happy to do it from Universitat Pompeu Fabra (UPF), the youngest university in Barcelona - counting only 20 years of life. We are proud to host TEI'13 from this small university (~13,000 undergraduate and graduate students), which is nonetheless (according to a recent official study by U. of Granada and U. of Zaragoza), the first university in Spain in percentage of scientific production per researcher.

Research paper thumbnail of Strictly Rhythm: Exploring the Effects of Identical Regions and Meter Induction in Rhythmic Similarity Perception

Lecture Notes in Computer Science, 2016

This paper is inspired in the ideas of rhythmical variation and evolution, which are connected to... more This paper is inspired in the ideas of rhythmical variation and evolution, which are connected to similarity and contrast. Two experiments on rhythm similarity are presented that examine the possible relations between objective metrics and human similarity ratings. We wanted to test the possible differences in similarity ratings when a beat was induced and when it was not. The experimental design is based on identical regions inserted in the rhythmic stimuli which are progressively shifted. Twentyone subjects participated in 2 experiments devised to calibrate the effect of identical regions and beat induction in similarity ratings. Results show that identical regions can influence similarity ratings more likely when there is not a meter induced. On the other hand, the induction of a pulse is prone to elicit an attention to coincidences between rhythms. It is also observed that coincidences in the first region of a rhythmic pattern have more importance than coincidences on other regions in order to be correlated to human similarity ratings. Practical consequences of these findings are discussed in the context of tools and agents for music creation.

Research paper thumbnail of Internet. Cap a Noves Formes d'entendre la Creació Musical?

Research paper thumbnail of 2003: Sonigraphical Instruments: From FMOL to the reacTable*

Current research in systematic musicology, 2017

This paper first introduces two previous software-based music instruments designed by the author,... more This paper first introduces two previous software-based music instruments designed by the author, and analyses the crucial importance of the visual feedback introduced by their interfaces. A quick taxonomy and analysis of the visual components in current trends of interactive music software is then proposed, before introducing the reacTable*, a new project that is currently under development. The reacTable* is a collaborative music instrument, aimed both at novices and advanced musicians, which employs computer vision and tangible interfaces technologies, and pushes further the visual feedback interface ideas and techniques aforementioned.

Research paper thumbnail of k-Best Unit Selection Strategies for Musical Concatenative Synthesis

Lecture Notes in Computer Science, 2018

Concatenative synthesis is a sample-based approach to sound creation used frequently in speech sy... more Concatenative synthesis is a sample-based approach to sound creation used frequently in speech synthesis and, increasingly, in musical contexts. Unit selection, a key component, is the process by which sounds are chosen from the corpus of samples. Hidden Markov Models are often chosen for this task, but one common criticism is its singular path output which is considered too restrictive when variations are desired. In this paper, we propose considering the problem in terms of k-Best path solving for generating alternative lists of candidate solutions and summarise our implementations along with some examples.

Research paper thumbnail of Real-Time Drum Accompaniment Using Transformer Architecture

Zenodo (CERN European Organization for Nuclear Research), Sep 17, 2022

This paper presents a real-time drum generation system capable of accompanying a human instrument... more This paper presents a real-time drum generation system capable of accompanying a human instrumentalist. The drum generation model is a transformer encoder trained to predict a short drum pattern given a reduced rhythmic representation. We demonstrate that with certain design considerations, the short drum pattern generator can be used as a real-time accompaniment in musical sessions lasting much longer than the duration of the training samples. A discussion on the potentials, limitations and possible future continuations of this work is provided.

Research paper thumbnail of Models to produce and share music over the Internet. The M.O.D.E.M. model

Research paper thumbnail of A Study of Tonal Practises In Electronic Dance Music

Research paper thumbnail of k-Best Hidden Markov Model Decoding for Unit Selection in Concatenative Sound Synthesis

Concatenative synthesis is a sample-based approach to sound creation used frequently in speech sy... more Concatenative synthesis is a sample-based approach to sound creation used frequently in speech synthesis and, increasingly, in musical contexts. Unit selection, a key component, is the process by which sounds are chosen from the corpus of samples. Hidden Markov Models are often chosen for this task, but one common criticism is its singular path output which is considered too restrictive when variations are desired. In this paper, we propose considering the problem in terms of k-Best path solving for generating alternative lists of candidate solutions and summarise our implementations along with some examples.

Research paper thumbnail of TEI 2010 Development Strategies for Tangible Interaction on Horizontal Surfaces

Research paper thumbnail of Sistemas acústicos y de tratamiento del sonido y el habla

Información del libro Sistemas acústicos y de tratamiento del sonido y el habla.

Research paper thumbnail of Evaluating rhythm similarity distances: The effect of inducing the beat

Research paper thumbnail of Drumming with style: From user needs to a working prototype

New Interfaces for Musical Expression, 2016

This paper documents and discusses the process of developing a generative drumming agent built fr... more This paper documents and discusses the process of developing a generative drumming agent built from the results of an extensive survey carried out with electronic music producers. Following the techniques of user-centered interaction design, an international group of beat producers was reviewed on the possibility of using AI algorithms to help them in the beat production work-flow. The results of these tests were used as design requirements for constructing a system that would indeed perform some tasks alongside the producer. The first results of this working prototype, a stylistic drum generator that creates new rhythmic patterns after being trained with a collection of drum tracks, are presented with a description of the system. Further stages of development and potential algorithms are also discussed.

Research paper thumbnail of Drum rhythm spaces: From polyphonic similarity to generative maps

Journal of New Music Research, Aug 24, 2020

This paper reports on the design of drum rhythm spaces as interactive bi-dimensional maps used fo... more This paper reports on the design of drum rhythm spaces as interactive bi-dimensional maps used for the visualization, retrieval and generation of drum patterns. We carry out two experiments exploring human processing of polyphonic drum patterns which conclude with a list of descriptors that significantly influence similarity sensations. These features are then used by a software system that is also described here to build rhythm spaces based on drum pattern collections. In the resulting spaces, patterns are organized by similarity, modelled according to human ratings. To enhance the functionality of rhythm spaces, a new algorithm for drum interpolation is introduced. This algorithm relates any point in a rhythm space with three drum patterns that bound it, going from a discrete space, that only retrieves patterns already in the collection, to a continuous generative space where each point in the space retrieves a specific pattern.

Research paper thumbnail of Past and Future of Physiological Computing and Creativity - An Underexplored and Promising Territory

Research paper thumbnail of SLFiducials

In this paper we introduce a new method for 6DoF marker tracking, specially designed for Microsof... more In this paper we introduce a new method for 6DoF marker tracking, specially designed for Microsoft SecondLight or any camera-based tabletop interface that is able to see objects through the surface. Our method is based on topological region adjacency for the identification of the markers, which are fitted into a squared shape for properly track the marker pose in the real world. We also describe the constraints imposed by the system which will determine the size and ID range of the new markers, and we finally evaluate the system.

Research paper thumbnail of Audio digital y MIDI

Research paper thumbnail of Target-based rhythmic pattern generation and variation with genetic algorithms

Composing drum patterns and musically developing them through repetition and variation is a typic... more Composing drum patterns and musically developing them through repetition and variation is a typical task in electronic music production. We propose a system that, given an input pattern, automatically creates related patterns using a genetic algorithm. Two distance measures (the Hamming distance and directed-swap distance) that relate to rhythmic similarity are shown to derive usable fitness functions for the algorithm. A software instrument in the Max for Live environment presents how this can be used in real musical applications. Finally, a user survey was carried out to examine and compare the effectiveness of the fitness metrics in determining rhythmic similarity as well as the usefulness of the instrument for musical creation.

Research paper thumbnail of Proceedings of the 7th International Conference on Tangible, Embedded and Embodied Interaction

Tangible and Embedded Interaction, Feb 10, 2013

A cohesive crack model is proposed to describe strain localization for the materials where strain... more A cohesive crack model is proposed to describe strain localization for the materials where strain-hardening is not prevailing over strain-softening (geomaterials, concrete-like materials, ceramics, etc.). Such a model is able to predict the size effects of fracture mechanics, i.e., the transition from ductile to brittle structure behaviour by increasing the size scale and keeping the geometrical shape unchanged. Whereas for Mode I, only untieing of the finite element nodes is applied to simulate crack growth, for Mixed Mode interelement crack propagation a topological variation is required at each step. In the case of four point shear testing, the load vs. deflection diagrams reveal snap-back instability for large sizes. By increasing the specimen sizes, such instability tends to reproduce the classical LEFM instability, predicted by the Maximum Circumferential Stress Criterion. Experimentally, the fracture toughness parameter of concrete appears to be unique and represented by the Mode I fracture energy Gp or the stress-intensity factor Kic even for Mixed Mode problems.

Research paper thumbnail of Prototyping interactions with Online Multimodal Repositories and Interactive Machine Learning

Interaction designers often use machine learning tools to generate intuitive mappings between com... more Interaction designers often use machine learning tools to generate intuitive mappings between complex inputs and outputs. These tools are usually trained live, which is not always feasible or practical. We combine RepoVizz, an online repository and visualizer for multimodal data, with a suite of Interactive Machine Learning tools, to demonstrate a technical solution for prototyping multimodal interactions that decouples the data acquisition step from the model training step. This way, different input data set-ups can be easily replicated, shared and experimented upon their capability to control complex output without the need to repeat the technical set-up.

Research paper thumbnail of Proceedings of the Sixth International Conference on Tangible, Embedded and Embodied Interaction

TEI is already in its seventh edition. This conference now counts with a solid and mature communi... more TEI is already in its seventh edition. This conference now counts with a solid and mature community, which is nonetheless growing every year. We are very happy to host this conference in Barcelona with one of the highest registrations ever, proving the high acceptance of the conference and its scientific value. We are also very happy to do it from Universitat Pompeu Fabra (UPF), the youngest university in Barcelona - counting only 20 years of life. We are proud to host TEI'13 from this small university (~13,000 undergraduate and graduate students), which is nonetheless (according to a recent official study by U. of Granada and U. of Zaragoza), the first university in Spain in percentage of scientific production per researcher.

Research paper thumbnail of Strictly Rhythm: Exploring the Effects of Identical Regions and Meter Induction in Rhythmic Similarity Perception

Lecture Notes in Computer Science, 2016

This paper is inspired in the ideas of rhythmical variation and evolution, which are connected to... more This paper is inspired in the ideas of rhythmical variation and evolution, which are connected to similarity and contrast. Two experiments on rhythm similarity are presented that examine the possible relations between objective metrics and human similarity ratings. We wanted to test the possible differences in similarity ratings when a beat was induced and when it was not. The experimental design is based on identical regions inserted in the rhythmic stimuli which are progressively shifted. Twentyone subjects participated in 2 experiments devised to calibrate the effect of identical regions and beat induction in similarity ratings. Results show that identical regions can influence similarity ratings more likely when there is not a meter induced. On the other hand, the induction of a pulse is prone to elicit an attention to coincidences between rhythms. It is also observed that coincidences in the first region of a rhythmic pattern have more importance than coincidences on other regions in order to be correlated to human similarity ratings. Practical consequences of these findings are discussed in the context of tools and agents for music creation.

Research paper thumbnail of Internet. Cap a Noves Formes d'entendre la Creació Musical?

Research paper thumbnail of 2003: Sonigraphical Instruments: From FMOL to the reacTable*

Current research in systematic musicology, 2017

This paper first introduces two previous software-based music instruments designed by the author,... more This paper first introduces two previous software-based music instruments designed by the author, and analyses the crucial importance of the visual feedback introduced by their interfaces. A quick taxonomy and analysis of the visual components in current trends of interactive music software is then proposed, before introducing the reacTable*, a new project that is currently under development. The reacTable* is a collaborative music instrument, aimed both at novices and advanced musicians, which employs computer vision and tangible interfaces technologies, and pushes further the visual feedback interface ideas and techniques aforementioned.

Research paper thumbnail of k-Best Unit Selection Strategies for Musical Concatenative Synthesis

Lecture Notes in Computer Science, 2018

Concatenative synthesis is a sample-based approach to sound creation used frequently in speech sy... more Concatenative synthesis is a sample-based approach to sound creation used frequently in speech synthesis and, increasingly, in musical contexts. Unit selection, a key component, is the process by which sounds are chosen from the corpus of samples. Hidden Markov Models are often chosen for this task, but one common criticism is its singular path output which is considered too restrictive when variations are desired. In this paper, we propose considering the problem in terms of k-Best path solving for generating alternative lists of candidate solutions and summarise our implementations along with some examples.

Research paper thumbnail of Real-Time Drum Accompaniment Using Transformer Architecture

Zenodo (CERN European Organization for Nuclear Research), Sep 17, 2022

This paper presents a real-time drum generation system capable of accompanying a human instrument... more This paper presents a real-time drum generation system capable of accompanying a human instrumentalist. The drum generation model is a transformer encoder trained to predict a short drum pattern given a reduced rhythmic representation. We demonstrate that with certain design considerations, the short drum pattern generator can be used as a real-time accompaniment in musical sessions lasting much longer than the duration of the training samples. A discussion on the potentials, limitations and possible future continuations of this work is provided.

Research paper thumbnail of Models to produce and share music over the Internet. The M.O.D.E.M. model

Research paper thumbnail of A Study of Tonal Practises In Electronic Dance Music

Research paper thumbnail of k-Best Hidden Markov Model Decoding for Unit Selection in Concatenative Sound Synthesis

Concatenative synthesis is a sample-based approach to sound creation used frequently in speech sy... more Concatenative synthesis is a sample-based approach to sound creation used frequently in speech synthesis and, increasingly, in musical contexts. Unit selection, a key component, is the process by which sounds are chosen from the corpus of samples. Hidden Markov Models are often chosen for this task, but one common criticism is its singular path output which is considered too restrictive when variations are desired. In this paper, we propose considering the problem in terms of k-Best path solving for generating alternative lists of candidate solutions and summarise our implementations along with some examples.

Research paper thumbnail of TEI 2010 Development Strategies for Tangible Interaction on Horizontal Surfaces

Research paper thumbnail of Sistemas acústicos y de tratamiento del sonido y el habla

Información del libro Sistemas acústicos y de tratamiento del sonido y el habla.

Research paper thumbnail of Evaluating rhythm similarity distances: The effect of inducing the beat

Research paper thumbnail of Drumming with style: From user needs to a working prototype

New Interfaces for Musical Expression, 2016

This paper documents and discusses the process of developing a generative drumming agent built fr... more This paper documents and discusses the process of developing a generative drumming agent built from the results of an extensive survey carried out with electronic music producers. Following the techniques of user-centered interaction design, an international group of beat producers was reviewed on the possibility of using AI algorithms to help them in the beat production work-flow. The results of these tests were used as design requirements for constructing a system that would indeed perform some tasks alongside the producer. The first results of this working prototype, a stylistic drum generator that creates new rhythmic patterns after being trained with a collection of drum tracks, are presented with a description of the system. Further stages of development and potential algorithms are also discussed.

Research paper thumbnail of Drum rhythm spaces: From polyphonic similarity to generative maps

Journal of New Music Research, Aug 24, 2020

This paper reports on the design of drum rhythm spaces as interactive bi-dimensional maps used fo... more This paper reports on the design of drum rhythm spaces as interactive bi-dimensional maps used for the visualization, retrieval and generation of drum patterns. We carry out two experiments exploring human processing of polyphonic drum patterns which conclude with a list of descriptors that significantly influence similarity sensations. These features are then used by a software system that is also described here to build rhythm spaces based on drum pattern collections. In the resulting spaces, patterns are organized by similarity, modelled according to human ratings. To enhance the functionality of rhythm spaces, a new algorithm for drum interpolation is introduced. This algorithm relates any point in a rhythm space with three drum patterns that bound it, going from a discrete space, that only retrieves patterns already in the collection, to a continuous generative space where each point in the space retrieves a specific pattern.

Research paper thumbnail of Past and Future of Physiological Computing and Creativity - An Underexplored and Promising Territory