In modern medicine, every cardiac assessment or respiratory check-up includes an audio auscultation during which one the medical specialist listens to sounds from the patient body with different tools (stethoscope, sonography). This shows how important sound analysis is for heart and lungs disease detection. During the IeBRI 2017 challenge, a database of 920 records acquired from 126 subject, was used to find a method that predicted if a respiratory cycle contains, or not, adventitious sounds like crackles, wheezes or both of them. The team which submits the best results reached around 50% of correct detection. Using a machine learning approach with a boosted decisional tree model and more audio features leads to the same results. A new approach consists in creating a new model at the patient level, which is able to decide if a patient sounds sick or not, by taking as input the predicted results of the first classification model. This new model permits to reach 85% of good predictions and could be used as a tool for helping doctors to make better diagnosis.

International audience

This article proposes to analyze the structural regularities from the audio and video streams of TV-programs and explore their potential for the classification of videos into program collections. Our approach is based on the spectral analysis of distance matrices representing the short-and long-term dependancies within the audio and visual modalities of a video. We propose to compare two videos by their respective spectral features. We appreciate the benefits brought by the two modalities on the performances in the context of a K-nearest neighbor classification, and we test our approach in the context of an unsupervised clustering algorithm. These evaluations are performed on two datasets of French and Italian TV programs.

This paper introduces a new variant of gameplay metrics that further develops a set of processes that expand user-centered game testing practices capable of quantifying user experiences. The key goal of the method presented here is to widen the appeal and application of game metrics within research relevant to, and representative of the wider field of game studies. In doing so, we acknowledge that the interests of this research community is often focused on player experience and performance with a broad range of off-the-shelf games that have already been released to the public. In order to be able to include any PC game system within research (or audiovideo stream, e.g., YouTube walkthroughs) our approach comprises of a postprocessing method for analyzing player performance. Through exploiting the audiovisual streams that are transmitted to the player, this approach processes content activated and generated during play and is therefore representative of individual player's encounters with specific games.

Today, technologies for information storage and transmission allow the creation and development of huge databases of multimedia content. Tools are needed to facilitate their access and browsing. In this context, this article focuses on the segmentation of a particular category of multimedia content, audio-visual musical streams, into music pieces. This category includes concert audio-video recordings, and sequences of music videos such as the ones found in musical TV channels. Current approaches consist in supervised clustering in a few audio classes (music, speech, noise), and, to our knowledge, no consistent evaluation has been performed yet in the case of audio-visual musical streams. In this paper, we aim at estimating the temporal boundaries of music pieces relying on the assumed homogeneity of their musical and visual properties. We consider an unsupervised approach based on the generalized likelihood ratio to evaluate the presence of statistical breakdowns of MFCCs, Chroma vectors, dominant Hue and Lightness over time. An evaluation of this approach on 15 manually annotated concert streams shows the advantage of combining tonal content features to timbral ones, and a modest impact from the joint use of visual features in boundary estimation.

Computations in most music retrieval systems strongly depend on the size of data compared. We propose to enhance performances of a music retrieval system, namely a harmonic similarity evaluation method, by first indexing relevant parts of music pieces. The indexing algorithm represents each audio piece exclusively by its major repetition, using harmonic descriptions and string matching techniques. Evaluations are performed in the context of a state-of-the-art retrieval method, namely cover songs identification, and results highlight the success of our indexing system in keeping similar results while yielding a substantial gain in computation time.

Query-by-tapping systems are content-based music retrieval systems that allow users to tap or clap in a microphone the rhythmic pattern of the melody requested. In this paper, a new query-by-tapping system is described. This system is based on adaptations of alignment algorithms successfully applied for melodic similarity estimation. A similarity score is computed according to the cost of the elementary operations necessary to transform the query into the musical piece tested. A new method for extracting the rhythmic pattern from audio signals is also presented. This method is based on the analysis of the variations of the kurtosis computed from the audio samples. Experiments performed on the MIREX 2008 database show that the alignment technique proposed performs better than the participating algorithms. They also confirm the interest of the kurtosis-based analysis method in the case of noisy percussive queries.

Retrieval systems for polyphonic music rely on the automatic estimation of similarity between two musical pieces. In the case of symbolic music, existing systems either consider a monophonic reduction based on melody or propose algorithms with high complexity. In this paper, we propose a new approach. Musical pieces are represented as a sequence of chords which are estimated from groups of notes sounding at the same time. A root and a mode are associated to each chord. Local alignment is then applied for estimating a similarity score between these sequences. Experiments performed on MIDI files collected on the Internet show that the system proposed allows the retrieval of different versions of the same song.

Existing symbolic music comparison systems generally consider monophonic music or monophonic reduction of polyphonic music. Adaptation of alignment algorithms to music leads to accurate systems, but their extensions to polyphonic music raise new problems. Indeed, a chord may match several consecutive notes, or the difference between two similar motifs may be a few swapped notes. Moreover, the substitution scores between chords are difficult to set up. In this paper, we propose a general framework for polyphonic music using the substitution score scheme set for monophonic music, which allows new operations by extending the operations proposed by Mongeau and Sankoff [15]. From a practical point of view, the limitation of chord sizes and the number of notes that can be merged consecutively lead to a complexity that remains quadratic.

Some natural sounds, such as speech parts can essentially be considered as noises. For instance, models suppose noisy parts of sounds as weak parts and apply basic approximations. But transformations such as time stretching do not preserve the noisy characteristics of sounds. Moreover, we show that those transformations introduce artificial intensity variations. In this paper we propose a spectral model for noise modeling which takes into account the statistical properties of such sounds. The analysis is based on the classical spectral models. The synthesis consists of randomly defining sinusoidal components. These components are then added using the adapted overlap-add method to keep statistical moments constant. Time scaling operations using this approach are described. Experiments on artificial sounds (filtered white noises) as well as natural sounds such as consonants and whispered vowels, show impressive enhancement in quality. Infinite time stretching transformations of such noises can be perfectly performed.