News and Events

Agustin Martorell defends his PhD thesis
19 Sep 2013

Agustín Martorell defends his PhD thesis entitled "Modelling tonal context dynamics by temporal multi-scale analysis" on Thursday 19th of September 2013 at 10:00h in room 55.410.

The jury members of the defense are: Petri Toiviainen (University of Jyväskylä), Geoffroy Peeters (IRCAM), Sergi Jordà (UPF).

Abstract: This work explores the multidimensional, ambiguous and temporal characteristics of tonality from a holistic perspective. The approach relies on interfacing pitch-spaces with time vs. time-scale descriptions. In this combined representation, the spatial and temporal hierarchies of tonality are evidenced simultaneously and in relation to each other. A visual exploration method is proposed for the analysis of tonal context in music works, using a simple model of tonal induction. A geometrical colouring solution, based on the topology of the pitch-space, approaches the perceptual correlation between the tonal properties and the visual representation. A relational taxonomy is proposed for describing tonal ambiguity, which leads to extending the method for the analysis of music based on tonal systems beyond the major-minor paradigm. Two perceptual studies are approached from this descriptive framework. The first study evidences the impact of time-scale in a simple model of tonal induction, and analyses the mathematical artefacts introduced by evaluations in scaled spaces. In the second study, a model of contextual instability is proposed and discussed in relation to the modelling of tonal tension. The analysis and representation methods are then generalised, through a set-class theoretical domain, in order to be applied with any pitch-based music.

16 Sep 2013 - 09:55 | view
6 PhD thesis defenses in 15 days!!!
In the next 15 days we have the defense of six PhD thesis done at the MTG!!!
 
September 6th - Dmitry Bogdanov: "From Music Similarity to Music recommendation: Computational Approaches based on Audio and Metadata Analysis"
September 9th - Ricard Marxer: "Audio Source Separation for Music in Low-latency and High-latency Scenarios" 
September 10th - Saso Musevic: "Non-stationary Sinusoidal Analysis" 
September 19th -  Agustin Martorell: "Modelling Tonal Context Dynamics by Temporal Multi-scale Analysis"
September 19th  - Jose R. Zapata: "Comparative Evaluation and Combination of Automatic Rhythm Description Systems"
September 20th - Justin Salamon: "Melody Extraction from Polyphonic Music Signals"
6 Sep 2013 - 17:27 | view
Saso Musevic defends his PhD thesis on September 10th
10 Sep 2013
Saso Musevic defends his PhD thesis entitled "Non-stationary sinusoidal analysis" on Tuesday 10th of September 2013 at 11:00h in room 55.309.

The jury members of the defense are: Axel Roebel (IRCAM), Emmanuel Vincent (INRIA), Marcelo Bertalmio (UPF).

Abstract: Many types of everyday signals fall into the non-stationary sinusoids category. A large family of such signals represent audio, including acoustic/electronic, pitched/transient instrument sounds, human speech/singing voice, and a mixture of all: music. Analysis of such signals has been in the focus of the research community for decades. The main reason for such intense focus is the wide applicability of the research achievements to medical, financial and optical applications, as well as radar/sonar signal processing and system analysis. Accurate estimation of sinusoidal parameters is one of the most common digital signal processing tasks and thus represents an indispensable building block of a wide variety of applications. Classic time-frequency transformations are appropriate only for signals with slowly varying amplitude and frequency content - an assumption often violated in practice. In such cases, reduced readability and the presence of artefacts represent a signi ficant problem. Time and frequency resolution cannot be increased arbitrarily due to the well known time-frequency resolution trade-o ff by Heisenberg. The main objective of this thesis is to revise and improve existing methods, and to propose several new approaches for the analysis of non-stationary sinusoids. This dissertation substantially contributes to the existing sinusoidal analysis algorithms: a) it critically evaluates and disseminates in great detail current analysis methods, b) provides signi ficant improvements for some of the most promising existing methods, c) proposes several new approaches for analysis of the existing sinusoidal models and d) proposes a very general and flexible sinusoidal model together with a fast, direct estimator.

6 Sep 2013 - 17:19 | view
Seminar by Axel Roebel and Emmanuel Vincent
9 Sep 2013

Axel Roebel from IRCAM and Emmanuel Vincent from Inria Nancy will give a seminar on Monday 9th of September 2013 at 16:00h in room 55.309

Axel will present an overview of the research carried out during the last five years at the Analysis-Synthesis team at IRCAM.
Emmanuel will introduce his recent work on symbolic music modeling with interpolated probabilistic models.

MORE INFORMATION:
Axel Roebel is the head of the Analysis-Synthesis team at IRCAM
Emmanuel Vincent is an experienced research scientist at Inria Nancy - Grand Est

6 Sep 2013 - 16:33 | view
Agreement for students exchange with IIT-Madras

The negotiation of the partnership agreement between the MTG and the Department of Science and Engineering at  IIT-Madras (Chennai, India) ended last July and it is related to the collaboration previously established between both groups over the performance of the CompMusic project. This project is funded by the European Research Council (ERC) and lead by Prof. Xavier Serra and it aims to advance in the automatic description of music by taking a culture specific approach. It focuses on five non occidental music traditions belonging to China, India, North Africa and Turkey. Two of them, the Hindustani and the Carnatic have the origin in the north of India and the south of India respectively.

According to the agreement clauses, both institutions will exchange students with the idea to keep a constant flux of visitors between them. Faculty members from both institutions have committed themselves to supervise the students that will participate in a research program of common interest for the two groups.

Student Exchange Conditions

This specific agreement may be also renewed and it will actually remain in effect until 2018. Moreover, it will cover the participation of students that are pursuing a research project as part of their degree (B. Tech, M. Tech, M. Sc., PhD or equivalent)
Each part agreed to accept up to three students per academic year and each participant may not spend more than one academic year in the exchange program. Although the host institution will reserve the right to make final judgment on the admission of the shortlisted candidates, following the agreement terms, the participant students will be selected by the home institution.

Next steps

The idea that MTG has is to replicate this exchange agreement with the different institutions taking part in the CompMusic project, such as the Indian Institute of Technology in Bombay or the Bahçeşehir Üniversitesi in Turkey.

The initiative promoted by MTG-UPF faculty and research members launches a new academic and research collaboration program that it will not just enhance the student exchange, but also the culture enrichment and the science knowledge in the music technology field.

6 Sep 2013 - 12:54 | view
Emilia Gómez, Perfecto Herrera and Paco Gómez, editors of a JNMR Special Issue

The goal of this special issue is to gather relevant, high-quality research on computational methods and applications in ethnomusicology. The papers included here deal with different musical facets such as pitch, pulse and tempo, and voice timbre. They address different musical repertoires, from Central-African to Basque folk music. They also cover a broad area: tools, including data collections, methodology and Ethnomusicology core-problems.

The list of articles include:

5 Sep 2013 - 09:47 | view
Ricard Marxer defends his PhD thesis on September 9th
9 Sep 2013

Ricard Marxer defends his PhD thesis entitled "Audio Source Separation for Music in Low-latency and High-latency Scenarios" on Monday 9th of September 2013 at 11:00h in room 55.309.

The jury members of the defense are: Axel Roebel (IRCAM), Emmanuel Vincent (INRIA), Oscar Camara (UPF).

Abstract: The source separation problem in digital signal processing consists in finding the original signals that were mixed together into a set of mixture signals. Solutions to this problem have been extensively studied for the speci fic case of musical signals, however their application to real-world practical situations remains infrequent. There are two main obstacles for their widespread adoption depending on the scenario. The main limitation in some cases is their high latency and computational requirements. In other cases the quality of the results is still unacceptable. There has been extensive work on improving the quality of music separation, but few studies have been devoted to the development of low-latency and low computational cost separation of monaural music signals. We propose speci fic methods to address these issues in each of these scenarios independently. First, we focus on methods with low computational cost and low latency. We propose the use of Tikhonov regularization as a method for spectrum decomposition in the low-latency context. We compare it to existing techniques in pitch and multipitch estimation and tracking tasks, which are a crucial step in many separation methods. We then use the proposed spectrum decomposition method in low-latency music separation tasks targeting singing voice, bass and drums. Second, we develop methods that achieve improved separation results with respect to existing state-of-the-art methods at the cost of greater computational cost and higher latency. We propose several high-latency and computationally complex methods that improve the separation of singing voice, by modeling components that are often not accounted for, such as breathiness and the consonants. Finally we explore the use of temporal correlations and human annotations to enhance the separation of drums and complex polyphonic music signals.

5 Sep 2013 - 09:26 | view
Dmitry Bogdanov defends his PhD thesis on September 6th
6 Sep 2013

Dmitry Bogdanov defends his PhD thesis entitled "From music similarity to music recommendation: Computational approaches based on audio and metadata analysis" on Friday 6th of September 2013 at 11:00h in room 55.309.

The jury members of the defense are: Markus Schedl (Johannes Kepler University), Josep Lluis Arcos (IIIA-CSIC), Emilia Gómez (UPF).

Abstract: The amount of music available digitally is overwhelmingly increasing. Vast amounts of music are available for listeners, and require automatic organization and filtering. In this context, user modeling, which consists in customization and adaptation of systems to the user's specific needs, is a challenging fundamental problem. A number of music applications are grounded on user modeling to provide users a personalized experience. In the present work we focus on user modeling for music recommendation, and propose a preference elicitation technique in conjunction with different recommendation approaches. We develop algorithms for computational understanding and visualization of music preferences. Our approaches employ algorithms from the fields of signal processing, information retrieval, machine learning, and are grounded in cross-disciplinary research on user behavior and music. Firstly, we consider a number of preference elicitation strategies, and propose a user model starting from an explicit set of music tracks provided by this user as evidence of his/her preferences. The proposed strategy provides a noise-free representation of music preferences. Secondly, we study approaches to music similarity, working solely on audio content. We propose a novel semantic measure which benefits from automatically inferred high-level description of music. Moreover, we complement it with low-level timbral, temporal, and tonal information and propose a hybrid measure. The proposed measures show significant improvement, compared to common music similarity measures, in objective and subjective evaluations. Thirdly, we propose distance-based and probabilistic recommendation approaches working with explicitly given preference examples. Both content-based and metadata-based approaches are considered. The proposed methods employ semantic and hybrid similarity measures as well as they build semantic probabilistic model of music preference. Further filtering by metadata is proposed to improve results of purely content-based recommenders. Moreover, we propose a lightweight approach working exclusively on editorial metadata. Human evaluations show that our approaches are well-suited for music discovery in the long tail, and are competitive with metadata-based industrial systems. Fourthly, to provide insights on the nature of music preferences, we create regression models explaining music preferences of our participants and demonstrate important predictors of their preference from both acoustical and semantic perspectives. The obtained results correlate with existing research on music cognition. Finally, we demonstrate a preference visualization approach which allows to enhance user experience in recommender systems.

3 Sep 2013 - 09:47 | view
Presentation of DREANSS dataset at DAFX'2013

The 16th International Conference on Digital Audio Effects DAFx 2013 is held in Maynooth, September 2-5 2013. MTG researchers Ricard Marxer and Jordi Janer participate with four papers related to different works on Audio Source Separation (voice breathiness, voice fricatives, bass and drums).

A new dataset of audio recordings manual annotations will be published during DAFX'13. DREANSS (DRum Event ANnotations for Source Separation) aims to help develop research in source separation methods for polyphonic audio music mixtures containing drums.

2 Sep 2013 - 10:05 | view
Participation to SSW8

Xavier Serra and Martí Umbert participate to the 8th ISCA Speech Synthesis Workshop (SSW8) that takes place in Barcelona from August 31st to  September 2nd, 2013. Xavier gives a keynote on "Singing voice synthesis in the context of music technology research" and Martí presents a paper on "Systematic database creation for expressive singing voice synthesis control".

  • Martí Umbert, Jordi Bonada, Merlijn Blaauw: "Systematic database creation for expressive singing voice synthesis control"
    Abstract: Systematic database creation for expressive singing voice synthesis control In the context of singing voice synthesis, the generation of the synthesizer controls is a key aspect to obtain expressive performances. In our case, we use a system that selects, transforms and concatenates units of short melodic contours from a recorded database. This paper proposes a systematic procedure for the creation of such database. The aim is to cover relevant style-dependent combinations of features such as note duration, pitch interval and note strength. The higher the percentage of covered combinations is, the less transformed the units will be in order to match a target score. At the same time, it is also important that units are musically meaningful according to the target style. In order to create a style-dependent database, the melodic combinations of features to cover are identified, statistically modeled and grouped by similarity. Then, short melodic exercises of four measures are created following a dynamic programming algorithm. The Viterbi cost functions deal with the statistically observed context transitions, harmony, position within the measure and readability. The final systematic score database is formed by the sequence of the obtained melodic exercises.
     
  • Xavier Serra: "Singing voice synthesis in the context of music technology research"
    Abstract: The synthesis of the singing voice has always been very much tied to speech synthesis. Since the initial work of Max Mathews with Kelly and Lochbaum at Bell Labs in the 1950s many engineers and musicians have explored the potential of speech processing techniques in music applications. After reviewing some of this history I will present the work done in my research group to develop synthesis engines that could sound as natural and expressive as a real singer, or choir, and whose inputs could be just the score and the lyrics of the song. Some of this research is being done in collaboration with Yamaha and has resulted in the Vocaloid software synthesizer. In the talk I want to make special emphasis on the specificities of the music context and thus on the technical requirements needed for the use of a synthesis techology in music applications.

 

30 Aug 2013 - 11:46 | view
intranet