The automatic classification of environmental noise sources from their acoustic signatures recorded at the microphone of a noise monitoring system (NMS) is an active subject of research nowadays. This paper shows how hidden Markov models (HMM’s) can be used to build an environmental noise recognition system based on a timefrequency analysis of the noise signal. The performance of the proposed H...

Speech detection is an important first step for audio analysis on media contents, whose goal is to discriminate the presence of speech from non-speech. It remains a challenge owing to various sound sources included in media audio. In this work, we present a novel audio feature extraction method to reflect the acoustic characteristic of the media audio in the timefrequency domain. Since the degr...

The Time-Frequency Renyi Entropy (TFRE) uses a timefrequency distribution (TFD) of a signal to provide a measure of the signal information content and complexity in the timefrequency (TF) plane. The concept is applied to the problem of newborn seizure detection using the HRV signal. This papers provides an experimental comparison of the performance of the TFRE obtained from selected TFDs, inclu...

This paper concerns the use of speech enhancement to improve automatic speech recognition (ASR) performance in noisy environments. Speech enhancement systems are usually designed separately from a back-end recognizer by optimizing the frontend parameters with signal-level criteria. Such a disjoint processing approach is not always useful for ASR. Indeed, timefrequency masking, which is widely u...

Speech and other natural sounds show high temporal correlation and smooth spectral evolution punctuated by a few, irregular and abrupt changes. In a conventional Hidden Markov Model (HMM), such structure is represented weakly and indirectly through transitions between explicit states representing ‘steps’ along such smooth changes. It would be more efficient and informative to model successive s...

Modeling high-order nonlinear systems is an important issue in audio signal processing. It may be employed in real-time emulation of analog nonlinear systems, such as guitar distortion and amplifiers or other vintage electronic audio systems. This paper proposes a new method for obtaining an economical black-box model of nonlinear systems using the swept-sine technique, which extracts the harmo...

In this work, a sort of binary wavelet packages with multi-scale are introduced, which are generalizations of multivariant wavelet packages. Finitely Supported wavelet bases for Sobolev spaces is researched. Steming from a pair of finitely supported refinale functions with multi-scaled dilation factor in space 2 2 ( ) L R satsfying a very mild condition, we provide a novel method for designing ...

How to collect information optimally by scheduling a limited number of sensors is one of the key technologies of safety monitoring for large mechanical equipment systems. In this paper, an improved model of Pulse Coupled Neural Network is used for the discretization algorithm of continuous features. And based on classificatory error parameter and approximate dependence degree in Variable Precis...

Distributions that are highly concentrated in the time-frequency plane are presented. Since the idea for these distributions originated from the Wigner representation in the quantum mechanics, a review of this representation is done in the first part of the paper. Abstracting the physical sense of the quantum mechanics representation, we defined the “pseudo quantum” signal representation. On th...

This paper presents the results of an experiment aimed to evaluate the quality of different audio restoration algorithms based on different Short Time Spectral Attenuation methods. To single out the best computational methodologies for audio restoration an experiment was made, implementing a software (in DirectX plug-in form) which uses different algorithms. The software, working in real time, ...