We present an object-oriented implementation for sinusoidal modelling based on the C++ Sound Object Library (SndObj). We outline the background to this analysis/synthesis technique and its inclusion in many well known methods of speech and music signal processing. Incorporation of such a well known technique into the SndObj library will enable the development of further audio processing techniques such as vocoding, time and pitch scaling and cross-synthesis on an object-oriented development platform.

Download

The computational power provided by current general purpose computers allows to undertake the implementation of low cost software-only real time audio processors. Unfortunately, computational power is not the only requirement for high demand applications. There are still important difficulties to overcome in other areas such as robustness and low latency. These difficulties have lead us to the development of Rappid, a development framework for C++ real-time high-demand audio processing applications. This paper describes the Rappid development framework. First of all we discuss the objectives we pursue with its development, and we give an overview to some other existent solutions before starting the actual description of the framework. We finally describe a first sound processing application which has been successfully developed with it.

Download

This paper proposes the sonification of the activity of a computer system that allows the user to monitor the basic performance parameters of the system, like CPU load, read and write activity of the hard disk or network traffic. Although, current computer systems still produce acoustic background noise, future and emerging computer systems will be more and more optimized with respect to their noise emission. In contrast to most of the concepts of auditory feedback, which present a particular sound as a feedback to a user’s command, the proposed feedback is mediated by the running computer system. The user’s interaction stimulates the system and hence the resulting feedback offers more realistic information about the current states of performance of the system. On the one hand the proposed sonification can mimic the acoustical behavior of operating components inside a computer system, while on the other hand, new qualities can be synthesized that enrich interaction with the device. Different forms of sound effects and generation for the proposed auditory feedback are realized to experiment with the usage in an environment of silent computer systems.

Download

This paper presents a new noise-shaping criterion. Based on the new criterion, we derive an efficient bit allocation method. The bit allocation method is applicable to the current audio standards like MPEG1 Layer 3 and MPEG4 AAC. The bit allocation method has gained a speed up for more than ten and has resulted in better quality over the traditional two nested loop method presented in ISO draft. The experiments illustrated the correction of the objective measurement criterion and the new allocation has shown the deterministic method instead of the iteration method to achieve the high allocation efficiency and best quality.

Download

Additive sinusoidal synthesis is a popular technique for applications like sound synthesis or very low bit rate parametric audio decoding. In this paper, different algorithms for the efficient synthesis of sinusoids on general purpose CPUs as found in today’s PCs are investigated. Fast algorithms for time domain synthesis of constant and linearly changing frequencies are presented and compared to frequency domain synthesis approaches. Execution time and accuracy (SNR) of the algorithms are reported for different CPU types. Finally, the algorithms are implemented in a fast MPEG-4 HILN parametric audio decoder in order to evaluate their performance in a real world application.

Download

This paper discusses the design and implementation of a scalable audio compression scheme that scales up from lossy to lossless compression. Scalable audio compression has been of interest in the audio compression community for some time, with the most obvious attempt at obtaining a solution coming in the form of the MPEG-4 standard [1]. At the same time the increase in bit rates in both mobile communications [2] and the internet’s broadband technology means that audio compression algorithms with higher bit rates than currently used, such as MPEG’s mp3 [1], can be employed to obtain higher quality. However, the new increased data rates are not necessarily constant, this is especially the case when considering the internet. As such, scalable schemes that can scale to lossless compression have become rather interesting from an application point of view. The scheme presented in this paper achieves lossless compression that is comparable with the state of the art whilst maintaining a scalable embedded bitstream.

Download

A current model of pitch perception is based on cochlear filtering followed by a periodicity detection. Such a computational model is implemented and then extended to characterise the sensory consonance of pitch intervals. A simple scalar measure of sensory consonance is developed, and to evaluate this perceptually related feature extraction the consonance is computed for musical intervals. The relation of consonance and dissonance to the psychoacoustic notions of roughness and critical bandwidth is discussed.

Download

Musical signals exhibit periodic temporal structure that create the sensation of rhythm. In order to model, analyze, and retrieve musical signals it is important to automatically extract rhythmic information. To somewhat simplify the problem, automatic algorithms typically only extract information about the main beat of the signal which can be loosely defined as the regular periodic sequence of pulses corresponding to where a human would tap his foot while listening to the music. In these algorithms, the beat is characterized by its frequency (tempo), phase (accent locations) and a confidence measure about its detection. The main focus of this paper is the concept of Beat Strength, which will be loosely defined as one rhythmic characteristic that could allow to discriminate between two pieces of music having the same tempo. Using this definition, we might say that a piece of Hard Rock has a higher beat strength than a piece of Classical Music at the same tempo. Characteristics related to Beat Strength have been implicitely used in automatic beat detection algorithms and shown to be as important as tempo information for music classification and retrieval. In the work presented in this paper, a user study exploring the perception of Beat Strength was conducted and the results were used to calibrate and explore automatic Beat Strength measures based on the calculation of Beat Histograms.

Download

Swim Swan is a multimedia work for clarinet, live electronics and computer graphics. The principles underlying the work are exposed and a set of sound-processing and synthesis algorithms that are used in this work are described. The issue of the control of the computer by the clarinet is emphasized. The artistic issues of such a work are addressed.

Download