Participation at the AES Audio for Games conference
Oscar Mayor participates at the AES 35th International Conference on Audio For Games that takes places in London from the 11th to the 13th of February 2009. Oscar is presenting one paper and one poster from work done at the MTG.
- Mayor, O., Bonada, J., Janer, J. KaleiVoiceCope: Voice Transformation from Interactive Installations to Video-Games.
- Mayor, O., Bonada, J., Loscos, A. Performance Analysis and Scoring of the Singing Voice.
12 Feb 2009 - 19:01
Seminar by Johan Sundberg on the singing voice
Johan Sundberg, from the Royal Institute of Technology of Sweden, will give a seminar on "The voice as a music instrument" on Thursday 12th of February at 15:00h in room 52.321 at the Roc Boronat building of the Comunication-Poblenou Campus.
By varying the air pressure in the lungs, the length and tension of the
vocal folds, the glottal adduction and the shape of the vocal tract the
human voice can be continuously varied along three perceptual dimensions:
loudness, pitch, and timbre. In this sense the human voice is more
flexible than other music instruments. Contributions to timbral variation
emanate from both the vibrating vocal folds and from the resonatory
properties of the vocal tract. The presentation will review some
investigations of how singers use these means for the purpose of vocal
economy and musical expressivity.
5 Feb 2009 - 16:37
Oscar Celma defends his PhD thesis on February 16th
Oscar Celma defends his PhD thesis entitled "Music Recommendation and Discovery in the Long Tail" on Monday 16th of February at 11:00h in room 52.223 at the Roc Boronat building of the Communication-Poblenou Campus.
The members of the jury's defense are: Ricardo Baeza-Yates (Yahoo! Research), Rafael Ramirez (UPF), Stephan Baumann (DFKI GMbH), Josep Lluis Arcos (IIIA-CSIC), Marc Torrens (Strands).
Thesis Abstract: Music consumption is biased towards a few popular artists. For instance, in 2007 only 1% of all digital tracks accounted for 80% of all sales. Similarly, 1,000 albums accounted for 50% of all album sales, and 80% of all albums sold were purchased less than 100 times. There is a need to assist people to filter, discover, personalise and recommend from the huge amount of music content available along the Long Tail.
Current music recommendation algorithms try to accurately predict what people demand to listen to. However, quite often these algorithms tend to recommend popular —or well–known to the user—music, decreasing the effectiveness of the recommendations. These approaches focus on improving the accuracy of the recommendations. That is, try to make accurate predictions about what a user could listen to, or buy next, independently of how useful to the user could be the provided recommendations.
In this Thesis we stress the importance of the user’s perceived quality of the recommendations.
We model the Long Tail curve of artist popularity to predict —potentially— interesting and unknown music, hidden in the tail of the popularity curve. Effective recommendation
systems should promote novel and relevant material (non–obvious recommendations),
taken primarily from the tail of a popularity distribution.
The main contributions of this Thesis are: (i) a novel network–based approach for recommender systems, based on the analysis of the item (or user) similarity graph, and the
popularity of the items, (ii) a user–centric evaluation that measures the user’s relevance
and novelty of the recommendations, and (iii) two prototype systems that implement the
ideas derived from the theoretical work. Our findings have significant implications for
recommender systems that assist users to explore the Long Tail, digging for content they
5 Feb 2009 - 13:21
El MTG col·labora amb el Programa Joves i Ciència de l'Obra Social de Caixa Catalunya
El MTG coordina el projecte "Els
Sons de la Natura" dins del Programa Joves i Ciència que l'Obra Social de Caixa Catalunya organitza a Planes de Son a l'estiu del 2009. Emilia Gomez i Jordi Bonada seran els professors i coordinadors d'aquest projecte.
Sons de la Natura és un projecte que
té com a objectiu explorar i entendre la natura a través dels seus
sons, fent servir les
tecnologies digitals relacionades amb el so i la música. Aquest
projecte està relacionat, i utilitzarà, la web Freesound.org;
infraestructura d'abast internacional desenvolupada per el MTG per a
experimentar amb sons i els seus usos de recerca i artístics.
El Programa Joves i Ciència està dirigit a joves que finalitzen 4t d'ESO i és realitza cada estiu a Planes de Son durant dues setmanes del mes de Juliol.
4 Feb 2009 - 13:06
Jordi Bonada defends his PhD thesis on February 13th
Jordi Bonada defends his PhD thesis entitled “Voice Processing and Synthesis by Performance Sampling and Spectral Models" on Friday 13th of February 2009 at 11:00h in room 104 of França Building.
The members of the jury's defense are: Marcelo Bertalmio (UPF), Sergi Jordà (UPF), Udo Zölzer (
Universität der Bundeswehr Hamburg, Germany), Johan Sundberg
(Royal Institute of Technology, Sweden) and Climent Nadeu (UPC).
Thesis Abstract: Singing voice is one of the most challenging musical instruments to model and imitate. Along several decades much research has been carried out to understand the mechanisms involved in singing voice production. In addition, from the very beginning of the sound synthesis techniques, singing has been one of the main targets to imitate and synthesize, and a large number of synthesizers have been created with that aim. The goal of this thesis is to build a singing voice synthesizer capable of reproducing the voice of a given singer, both in terms of expression and timbre, sounding natural and realistic, and whose inputs would be just the score and the lyrics of a song. This is a very difficult goal, and in this dissertation we discuss the key aspects of our proposed approach and identify the open issues that still need to be tackled.
This dissertation substantially contributes to the field of singing voice synthesis: a) it critically discusses spectral processing techniques in the context of singing voice modeling, and provides significant improvements to the current state of the art; b) it applies the proposed techniques to other application contexts such as real-time voice transformations, museum installations or video games; c) it develops the concept of synthesis based on performance sampling as a way to model the sonic space produced by a performer with an instrument, focusing on the specific case of the singing voice; d) it proposes and implements a complete framework for singing voice synthesis; e) it explores the sonic space of the singing voice and proposes a procedure to model it; f) it discusses the issues involved in the creation of the synthesizer‟s database and provide tools to automate its generation; g) it performs a qualitative evaluation of the synthesis results, comparing those to the state of the art and to real singer performance; h) it implements all the research results into an optimized software application for singing voice analysis, modeling, transformation and synthesis, including tools for database creation; i) a significant part of this research has been incorporated to a commercial singing voice software by Yamaha Corp.
2 Feb 2009 - 16:05
PhD positions at the MTG to start in the Fall 2009
The Music Technology Group of the Universitat Pompeu Fabra in Barcelona is opening 4 funded PhD positions to start in the Fall of 2009. The candidates have to apply before May 15th to the PhD program of the Department of Information and Communication Technologies of the UPF.
The MTG carries out research in topics such as as sound processing and synthesis; music content description;
interactive music systems; computational models of perceptual and music
cognition; and the technologies related with music social networks. The accepted PhD students will carry out research
within one of the projects of the MTG and thus the candidates have to
demonstrate a research record and make a research proposal related to
one of these topics. Depending on his/her academic background a
candidate may be asked to apply to the Master in Sound and Music Computing before being considered for the PhD.
PhD candidates should first contact Prof. Xavier Serra in order to get an initial approval of the research proposal to be submitted.
21 Jan 2009 - 10:15
Seminar by Daniel Erro on voice conversion
Next thursday, January 22nd 2008, Daniel Erro will give a seminar with the title "Voice conversion: state of the art, recent
improvements and remaining challenges" at 3pm in the França Auditorium.
Voice conversion, which consists of modifying the
voice of a speaker to be perceived as that of a different specific speaker, is
probably the most complex type of voice transformation. Its development opens
the door to very interesting applications: building multi-speaker or
multi-emotional speech synthesis systems, creating special voices or virtual
clones of famous people for videogames, films, toys, chat rooms, speaking aids
for people suffering from voice pathologies, etc. During the seminar, I will
give an overview of the voice conversion world. I will talk about the techniques
applied by state-of-the-art systems and I will focus mainly on the work carried
out at UPC to overcome some of the traditional limitations of voice conversion:
the quality degradation and the lack of flexibility in terms of training
requirements. Finally, I will list some of the future challenges for researchers
in this area and I will show some preliminary results of my current work at the
Aholab group (UPV/EHU).
Erro received the Telecommunication
Engineering degree from the Public University of Navarra (Pamplona, Spain, 2003)
and the PhD degree from the UPC (Barcelona) in june 2008. He is currently
working as a post-doctoral researcher at the Aholab group (University of the
Basque Country, Bilbao). He is interested in everything related to voice
transformation and conversion, signal models for speech and music waveform
analysis, modification and reconstruction, and speech
19 Jan 2009 - 09:41
LoopMash by Steinberg, developed in collaboration with the MTG, is announced
At the NAMM show, one of the world’s largest music trade shows, Steinberg Media Technologies GmbH has announced the release of Cubase 5, the newest version of its well known music production software. In cooperation with Yamaha Corporation, the Music Technology Group (MTG) of the Universitat Pompeu Fabra in Barcelona played an important role in the research and development that led to one of its new features, LoopMash, a powerful creative tool for slicing and instant re-assembling of sound material.
LoopMash originates from a joint research project between MTG and
Yamaha, a project that aimed at developing the next generation of
music production tools. In particular it focused on the manipulation
of sounds in a way similar to the creation of mosaics, a process that
is now known as audio mosaicing. With the techniques developed,
existing sound material can be recycled and recomposed using
computational tools; first by automatically extracting sound snippets
from existing music, then analyzing the similarity between them, and
finally recombining the snippets in musically relevant ways.
The outcome of the project was a software prototype that used this
technique as a musical instrument, remixing the snippets intelligently
and automatically, typically from drum-loops, under interactive
control of the musician. This prototype showed that potential of these
techniques for making music based on reusing existing sound material
and thus giving deejays and contemporary electronic musicians a new
level of control.
In a team effort, MTG, Steinberg and Yamaha successfully have converted the
resulting prototype into a real product. LoopMash is now a VST3
Instrument, integrated into Cubase, and thus adding to Cubase a unique
and innovative way of creating music.
16 Jan 2009 - 11:54
Talk by H. Purwins in Munich
Hendrik Purwins gave a talk on "Time series analysis of sound and music signal" at the Stochastic Approaches to Complexity workshop that took place at the Center for Nonlinear Science of the University of Munich from January 8th to the 10th 2009.
12 Jan 2009 - 16:13
Graduate programs on SMC for 2009-2010
The registration for the official Master in Sound and Music Computing 2009-2010 is
now open and the first registration period ends on January 30th, 2009 (check application calendar
part from the SMC Master, in the academic year 2009-10 the
ESMUC and the UPF will jointly offer specialized professional programs in
the area of Sound and Music Computing. Specifically in the period
September to December 2009 an intensive course on Music Production Technologies and Techniques
will be offered and another one on Interactive Music Systems Design
will be offered from January to April 2010. The official information about these courses should be available in January 2009.
22 Dec 2008 - 18:23