Tag Archives: music information retrieval

New IEEE/ACM TASP paper on multi-feature beat tracking

Multi-feature beat tracking Our article on multi-feature beat tracking for the IEEE/ACM Transactions on Audio, Speech and Signal Processing is now available online! This is a work carried leaded by Jose Ricardo Zapata for his PhD thesis in collaboration with Mathew Davies from the SMC group in Porto, based on the idea of combining different experts, represented by periodicity from different onset detection functions, for beat estimation. This is a simple and clever idea, already used to combine different beat tracking algorithms and evaluate the difficulty of the task, that has been integrated in a different method.

Zapata, J. R., Davies M. E. P., & Gómez E. (2014).  Multi-feature beat tracking. IEEE/ACM Transactions on Audio, Speech, and Language Processing. 22(4), 816 – 825. RTF, Tagged, XML, BibTex, Google Scholar

Abstract:

A recent trend in the field of beat tracking for musical audio signals has been to explore techniques for measuring the level of agreement and disagreement between a committee of beat tracking algorithms. By using beat tracking evaluation methods to compare all pairwise combinations of beat tracker outputs, it has been shown that selecting the beat tracker which most agrees with the remainder of the committee, on a song-by-song basis, leads to improved performance which surpasses the accuracy of any individual beat tracker used on its own. In this paper we extend this idea towards presenting a single, standalone beat tracking solution which can exploit the benefit of mutual agreement without the need to run multiple separate beat tracking algorithms. In contrast to existing work, we re-cast the problem as one of selecting between the beat outputs resulting from a single beat tracking model with multiple, diverse input features. Through extended evaluation on a large annotated database, we show that our multi-feature beat tracker can outperform the state of the art, and thereby demonstrate that there is sufficient diversity in input features for beat tracking, without the need for multiple tracking models.

Leave a comment

27/02/2014 · 15:53

IEEE Signal Processing Magazine Melody Extraction Review published online

Our review article on melody extraction algorithms for the IEEE Signal Processing Magazine is finally available online! The printed edition will be coming out in March 2014.

I believe (not just as I am a co-author!) that it will become a key reference in the Music Information Retrieval area and beyond, as it provides a very nice overview of approaches, challenges and applications for melody extraction from polyphonic music signals. Justin Salamon has been the main author (congratulations, Justin!) and the paper has benefit from the contribution of two key experts: Gaël Richard, and Dan Ellis, with who I had the chance to collaborate on a previous comparative study on melody extraction published at IEEE TASP (128 citations according to google scholar).

Finally, I like very much this kind of tutorial papers providing a comprehensive introduction to a given topic and with a very attractive design. I hope you will enjoy it!

Image
J. Salamon, E. Gómez, D. P. W. Ellis and G. Richard, “Melody Extraction from Polyphonic Music Signals: Approaches, Applications and Challenges“, IEEE Signal Processing Magazine, 31(2):118-134, Mar. 2014.

Abstract—Melody extraction algorithms aim to produce a sequence of frequency values corresponding to the pitch of the dominant melody from a musical recording. Over the past decade melody extraction has emerged as an active research topic, comprising a large variety of proposed algorithms spanning a wide range of techniques. This article provides an overview of these techniques, the applications for which melody extraction is useful, and the challenges that remain. We start with a discussion of ‘melody’ from both musical and signal processing perspectives, and provide a case study which interprets the output of a melody extraction algorithm for specific excerpts. We then provide a comprehensive comparative analysis of melody extraction algorithms based on the results of an international evaluation campaign. We discuss issues of algorithm design, evaluation and applications which build upon melody extraction. Finally, we discuss some of the remaining challenges in melody extraction research in terms of algorithmic performance, development, and evaluation methodology.

For further information about this article please visit Justin Salamon’s research page.

Leave a comment

Filed under publications, research, teaching

Video of my keynote talk at the 3rd International Workshop on Folk Music Analysis 2013

This is the video of my keynote talk at FMA 2013, titled “Towards Computer-Assisted Transcription and Description of Music Recordings”. This is the abstract of the talk. I hope you will like it!

Automatic transcription, i.e. computing a symbolic musical representation from a music recording, is one of the main research challenges in the field of sound and music computing. For monophonic music material the obtained transcription is a single musical line, usually a melody, and in polyphonic music there is an interest in transcribing the predominant melodic line. In addition to transcribing, current technologies are able to extract other musical descriptions related to tonality, rhythm or instrumentation from music recordings. Automatic description could potentially complement traditional methodologies for music analysis.

In this talk I present the state-of-the art on automatic transcription and description of music audio signals. I illustrate it with our own research on tonality estimation, melodic transcription and rhythmic characterization. I show that, although current research is promising, current algorithms are still limited in accuracy and there is a semantic gap between automatic feature extractors and expert analyses.
Moreover, I present some strategies to address these challenges by developing methods adapted to different repertoire and defining strategies to integrate expert knowledge into computational models, as a way to build systems following a “computer-assisted” paradigm.

Leave a comment

12/07/2013 · 12:56

Paper on “My musical avatar”

We feel ourselves identified with the type of music we like and we sometimes use music to define our personality. I guess one of the questions I ask to any new person I know is “whad kind of music do you listen to?”.

During the last few years, I have been taking part in a research project, where the main goal is to visualize one’s musical preferences, “The Musical Avatar“. The idea behind is to use computational tools to automatically describe your music (in audio format) in terms of melody, instrumentation, rhythm, etc and use this information to build an iconic representation of one’s musical preferences and to recommend you new music. All the system is only based on content description, i.e. on the signal itself and not on information about the music (context) as found on web sites, etc. And it works! 🙂

We finally published a paper describing the technology behind and its scientific evaluation   at Information Processing & Management journal. This is the complete reference:

Dmitry Bogdanov, Martín Haro, Ferdinand Fuhrmann, Anna Xambó, Emilia Gómez, Perfecto Herrera Semantic audio content-based music recommendation and visualization based on user preference examples. Information Processing & Management
Volume 49, Issue 1, pp. 13-33, January 2013

There is much to improve, but you can see my musical avatar below. Can you guess how my favorite music sounds like? You can of course build yours from your last-FM profile here.

Emilia's musical avatar

My automatically generated musical avatar

Highlights

► We propose preference elicitation technique based on explicit preference examples. ► We study audio-based approaches to music recommendation and preference visualization. ► Approaches based on semantics inferred from audio surpass low-level timbre methods. ► Such approaches are close to metadata-based system being suitable for music discovery. ► Proposed visualization captures the core musical preferences of the participants.

Leave a comment

Filed under projects, research

Computational Ethnomusicology and FMA (3rd IW on Folk Music Analysis)

Over the last few years, there has been an increasing interest in the study of music from different traditions from a computational perspective. Researchers with interests in this area have been meeting at the ISMIR conferences and communicate through an interest group in computational ethnomusicology, ethnocomp.

My interests started in 2008, with a study about how tonal features, extracted from music audio signals, can be useful to automatically organize music recordings from different traditions. It basically consisted on characterizing the scale by means of high-resolution HPCP features and combining these features with timbre and rhythm descriptors. As a result, we established some relationships between audio features and geography in our ISMIR2009 paper on Music and geography: content description of musical audio from different parts of the world. After that, I got interested in MIR and Flamenco music, and I have been working in a system for the automatic transcription of flamenco singing, thanks to the COFLA project. This is a challenging task, that will require a dedicated post!

ethnocomp has always been a small community, and two years ago we had the first event devoted to this research area, the first Folk Music Analysis (FMA) workshop that took place in Athens, Greece. Last year I had the chance of co-organizing the 2nd FMA in Seville, my home town, which was jointly organized with a conference on flamenco research. At the last ISMIR in Porto, we could see an increasing interest in this small field, and there was a large number of people attending the ethnocomp ‘dinner?. Moreover, at my research group, my boss Xavier Serra is leading an ERC grant dealing with MIR and traditional music, compmusic. I am very happy that this field gets more attention, and that we address the fact that all our technology has been designed for Western popular music. There is much work to do to develop culture-specific or culture-aware tools.

I then hope that this year’s FMA, which will take place in Amsterdam, will be a success! I am sure it will be a truly interdisciplinary event, gathering people from ethnomusicology, music performance and music information retrieval.

Topics include:
– Computational ethnomusicology
– Retrieval systems for non-western and folk musics
– New methods for music transcription
– Formalization of musical data
– Folk music classification systems
– Models of oral transmission of music
– Cognitive modelling of music
– Aesthetics and related philosophical issues
– Methodological issues
– Representational issues and models
– Audio and symbolic representations
– Formal and computational music analysis

Important dates:
3 February 2013: Deadline for abstract submissions
10 March 2013: Notification of acceptance/rejection of submissions
5 May 2013: Deadline for submission of revised abstracts or full papers
6 and 7 June: Workshop

Don’t miss it!!!!

Leave a comment

Filed under CFP, events, research

HPCP pluging available for free download

Image

We finally managed to share a simple version of our algorithm for chroma feature extraction (Harmonic Pitch Class Profile) with the research community by means of a vamp plugin. It’s currently available for windows, but we hope it will be soon available for MacOS and LINUX. You can find it here.

I am very happy for the success we had with the MELODIA plugin by Justin and I hope people will find this one interesting, even if the algorithm is from 2006!

The HPCP is an approach for chroma feature extraction. It provides a frame representation of the relative intensity of each pitch-class within an octave. I developed it as part of my PhD thesis and it has been extensively used for different Music Information Retrieval applications such as key and chord estimation, cover version identification, music structure analysis, classification and recommendation.

Leave a comment

19/10/2012 · 13:21

IEETASP paper online

Our paper on “Melody Extraction from Polyphonic Music Signals using Pitch Contour Characteristics”  is online! Congratulations to Justin!

Leave a comment

Filed under research