Search

Apart from speech and music, general sound can also carry relevant information. This is, however, a considerably less researched field up to-date. Most prominent in this area are the tasks of acoustic event detection and classification that can be subsumed under the area of computational auditory scene analysis. Fields of application include media retrieval including affective content analysis or human-machine and human-robot interaction, animal vocalisation recognition, and monitoring of industrial processes. Here, three applications in real-life Intelligent Sound Analysis are given from the work of the author: audio-based animal recognition, acoustic event classification, and prediction of emotion as induced in sound listeners. In particular, weakly supervised learning techniques are presented to cope with the typical label-sparseness in this field.

This study uses Genetic Programming (GP) in developing a classifier to distinguish between five musical instruments. Using only simple arithmetic and boolean operators with 95 features as terminals, a program is developed that can classify 300 unseen samples with an accuracy of 94%. The experiment is then run again using only 14 of the most often chosen features. Limiting the features in this way raised the best classification to 94.3% and the average accuracy from 68.2% to 75.67%. This demonstrates that not only can GP be used to create a classifier but it can be used to determine the best features to choose for accurate musical instrument classification, giving an insight into timbre.

One of the most common applications of snore sound analysis is to find out its relationship with obstructive sleep apnoea (OSA). The present work is aimed at developing a method for processing full night snore recordings obtained from a simple recording system that can be used both for clinical and home application. Automatic extraction of snore sounds only could in fact allow for saving much time usually required for manual analysis

Apnea is considered one of the major sleep disorders with great accession in population and significant impact on patient’s health. Symptoms include disruption of oxygenation, snoring, choking sensations, apneic episodes, poor concentration, memory loss, and daytime somnolence. Diagnosis of apnea involves monitoring patient’s biosignals and breath during sleep in specialized clinics requiring expensive equipment and technical personnel. This paper discusses the design and technical details of a platform capable for preliminary detection of sleep apnea at patient’s home utilizing snore analysis.

We address the problem of analyzing programs such as J2ME midlets for mobile devices, where a central correctness requirement concerns confidentiality of data that the user wants to keep secret. Existing software model checking tools analyze individual program executions, and are not applicable to checking confidentiality properties that require reasoning about equivalence among executions. We develop an automated analysis technique for such properties. We show that both over- and under- approximation is needed for sound analysis. Given a program and a confidentiality requirement, our technique produces a formula that is satisfiable if the requirement holds. We evaluate the approach by analyzing bytecode of a set of Java (J2ME) methods.

Snoring is the earliest and the most frequent symptom of Obstructive Sleep Apnea (OSA), which is a serious disease with a high community prevalence rate. The standard method of OSA diagnosis requires an overnight stay in a sleep lab, connected to over 15 channels of contact-measurements. There is evidence suggesting that Snore-Related-Sounds (SRS) carry sufficient information to diagnose OSA. Snore-based technology opens up opportunities for community-screening devices that do not depend on contact instrumentation. A new technique of diagnosing OSA based solely on multi-parametric snore sound analysis is presented in this paper. The method comprises of a logistic regression model fed with a range of snore parameters derived from its features, the pitch and the Total Airways Response (TAR) estimated using a Higher Order Statistics (HOS) based algorithm. The model was developed and its performance validated on a clinical database consisting of overnight snoring sounds of 41 subjects. Leaveone-out technique was used for validating the model. The validation process achieved 89.3% sensitivity with 92.3% specificity (area under the Receiver Operating Characteristic (ROC) curve was 0.96) in classifying the data sets into the two groups OSA and non-OSA. These results are superior to the existing results and unequivocally illustrate the feasibility of developing a snore-based OSA screening device.

The analysis of sounds generated in close proximity of a subject can often indicate emergency events like falls, pain and other distress situations. This paper presents a system for collecting and analyzing sounds and speech expressions utilizing on-body sensors and advanced classification techniques for emergency events detection. A variety of popular classification and meta-classification algorithms have been evaluated and the corresponding results are presented.

Software engineering requires creativity, thorough design and analysis, and sound design decisions. Design decisions often have tradeoffs and implications associated with them. Therefore, it is important that design decisions are based on sound analysis. With respect to embedded systems, key drivers are often performance and cost. Thus the purpose of this paper is to describe an approach to aid in the design decision process on cost and performance tradeoffs for embedded systems. Specifically, it presents a model-driven approach to understand and communicate the performance-cost tradeoff.

Our work is within the framework of studying and implementing a sound analysis system in a telemedicine project. The task of this system is to detect situations of distress in a patient’s room based on sound analysis. If such a situation is detected, an alarm will be automatically sent to the medical centre. In this paper we present our works on building domain ontology of such situations. They gather abstract concepts of sounds and these concepts, along with their properties and instances, are represented by a neural network. The ontology-based classifier uses outputs of networks to identify classes of audio scenes. The system is tested with a database extracted from films.

Sound synthesis using granular (or micro-sonic) techniques offers very attractive possibilities for creating musical sounds. We present a state of the researches conducted at GMEM for an integrated microsound synthesis system. This system includes at once synthesis generators and synthesis control programs, both gestures and sound analysis based control.

Our work aims at certifying that all the executions of several collaborating synchronous systems in a realistic environment follow a given specification. In order to analyze the numerous executions that may happen while considering a set of synchronous systems whose clocks are non-perfect and that communicate through non-instantaneous channels, we define two new abstract domains. The Changes counting domain and the Integral bounding domain gap the imprecisions of the previously defined Constraint domain that occur because of these hardware imprecisions. We define a reduced product between these domains that allows a much more precise though sound analysis than the three analyses that may have been defined in each domain.

Our work is within the framework of studying and implementing a sound analysis system in a telemedicine project. The task of this system is to detect situations of distress in a patient’s room based sound analysis. In this paper we present our works on building domain ontologies of such situations. They gather abstract concepts of sounds and these concepts, along with their properties and instances, are represented by a neural network. The ontology-based classifer uses outputs of networks to identify classes of audio scenes. The system is tested with a database extracted from films.

A sound analysis of microstructures relies on correct sampling and on the right choice of the direction in which thin sections are cut from samples. This chapter discusses the steps of sample collection; choice of sectioning plane, and problems involved in the interpretation of three-dimensional structures from two-dimensional sections.

Flow cytometry is a very fast optical (spectroscopic) analysis of cells on a cell-by-cell basis. The measurement, data acquisition and analysis were made possible by an innovative application of optical spectroscopy and high tech electronics, used originally in nuclear physics. A unique characteristic of the method is the analysis of individual cells at such a high speed that biological variations and their distribution over a conveniently large cell population are accessible to statistically sound analysis within a time scale of minutes among practically physiological conditions of the cells (Shapiro 1985; Ormerod 1994).