T. Jehan

Downbeat prediction by listening and learning

IEEE Workshop on Applications of Signal Processing to Audio and Acoustics, 2005., 2000

The perception of downbeat is not yet very well understood, although it generally "feels" natural... more The perception of downbeat is not yet very well understood, although it generally "feels" natural and intuitive to music listeners. We propose an unbiased and predictive modeling strategy of downbeat by combining psychoacoustic models of music listening with time-lag embedded segment learning. The model is causal, tempo independent, and could inform in a top-down manner a bottom-up beat tracker. Results with particularly complex music examples are presented.

Download

Hierarchical multi-class self similarities

IEEE Workshop on Applications of Signal Processing to Audio and Acoustics, 2005., 2005

ABSTRACT The music similarity question is yet to be defined. From an objective signal point of vi... more ABSTRACT The music similarity question is yet to be defined. From an objective signal point of view, at least three classes of similarities may be considered: pitch, rhythm, and timbre. Within each class, several scales of analysis may be evaluated: segment, beat, pattern, section. We propose a recursive approach to this hierarchical and perceptual similarity problem. Our musically meaningful set of measurements is computed from the ground up via dynamic programming. This may be regarded as an analysis foundation for content-based retrieval, song similarities, or manipulation systems.

Method and apparatus for automatically segueing between audio tracks

Perceptual segment clustering for music description and time-axis redundancy cancellation

Repeating sounds and patterns are widely exploited throughout music. However, although analysis a... more Repeating sounds and patterns are widely exploited throughout music. However, although analysis and music information retrieval applications are often concerned with processing speed and music description, they typically discard the benefits of sound redundancy cancellation. We propose a perceptually grounded model for describing music as a sequence of labeled sound segments, for reducing data complexity, and for compressing audio.

Download

Perceptual synthesis engine: an audio-driven timbre generator

... The fact that this is all processed on the By with little delay between bow-stroke and sound ... more

Musical signal parameter estimation

Creating music by listening

... From the space that makes any musical sense, an even smaller fraction of it is perceived as u... more

Creating music by concatenative synthesis

Interacting with music in a social setting

CHI '00 extended abstracts on Human factors in computing systems - CHI '00, 2000

This paper describes the design of a new system for interacting with music in a social setting. M... more This paper describes the design of a new system for interacting with music in a social setting. MusiCocktail allows users to influence certain parameters of a pre-composed and pre-recorded piece of music in the way they mix their beverages at a social gathering. This new form of interaction with music enables group participation in the creation of a rich musical

Convex non-negative matrix factorization for automatic music structure identification

2013 IEEE International Conference on Acoustics, Speech and Signal Processing, 2013

ABSTRACT We propose a novel and fast approach to discover structure in western popular music by u... more ABSTRACT We propose a novel and fast approach to discover structure in western popular music by using a specific type of matrix factorization that adds a convex constrain to obtain a decomposition that can be interpreted as a set of weighted cluster centroids. We show that these centroids capture the different sections of a musical piece (e.g. verse, chorus) in a more consistent and efficient way than classic non-negative matrix factorization. This technique is capable of identifying the boundaries of the sections and then grouping them into different clusters. Additionally, we evaluate this method on two different datasets and show that it is competitive compared to other music segmentation techniques, outperforming other matrix factorization methods.

The Echo Nest Analyzer Documentation

Analyzer documentation

MIREX 2014 Entry: Convex Non-negative Matrix Factorization

Perceptual Analysis of the F-Measure for Evaluating Section Boundaries in Music

by T. Jehan and Morwaread M Farbood

Music retrieval from everything

by T. Jehan and Paul Lamere

Proceedings of the international conference on Multimedia information retrieval - MIR '10, 2010

The Echo Nest Corporation, a music intelligence platform company based on years of music retrieva... more The Echo Nest Corporation, a music intelligence platform company based on years of music retrieval research and development, provides a number of retrieval, search and in- teractivity tools to music stores, social networks, musicians, developers and labels around the world. Our platform1 consists of a series of heterogeneous data types: text data from web crawls, audio signal processing and machine

Cati dance

by T. Jehan and C. Boulanger

Proceedings of the SIGGRAPH 2003 conference on Sketches & applications in conjunction with the 30th annual conference on Computer graphics and interactive techniques - GRAPH '03, 2003

ABSTRACT This sketch presents a real-time system that aims to bridge a gap between machine listen... more ABSTRACT This sketch presents a real-time system that aims to bridge a gap between machine listening technology, and self-editing, self-synchronizing video: a movie organizes itself from &quot;listening&quot; to music. In our current demonstration, a series of short video clips of &quot;Cati&quot; dancing, originally shot at different tempi, are arbitrarily sequenced, always in sync with the analyzed beat, i.e., if the music slows down, the dance slows down accordingly. When no beat is found, e.g., the music is mellow, then Cati stops dancing and waits, apparently bored.

Sparkler: An audio-driven interactive live computer performance for symphony orchestra

This paper describes the design, implementation, and application of an audio-driven computer-base... more This paper describes the design, implementation, and application of an audio-driven computer-based system for live and interactive performance with an orchestra. We begin by describing the compositional and musical goals, emphasizing the electronics aspect. We then discuss the challenges of working with an orchestra and integrating the electronics in a concert hall. We describe the hardware setup and then detail the software implementation, from audio analysis and "perceptual parameter" estimation to the generative algorithm.

Download

An audio-driven, spectral analysis-based, perceptual synthesis engine

by B. Schoner and T. Jehan

Determining the similarity of music using cultural and acoustic information

An audio-driven perceptually meaningful timbre synthesizer

Analysis

... component. In order to extract the instantaneous loudness of the signal, the power-spectrum b... more

Uploads

Papers by T. Jehan

Log In