Download Digital Synthesis By Plug-In Method In Java Media Framework Environment
This paper deals with the implementation of real-time digital musical sound synthesizers by the Plug-In method in the Sun Microsystems Java Media Framework environment. This environment use the Plug-In technology as well as the DirectX or VST environments, but the implementation methods are different.
Download Complex domain onset detection for musical signals
We present a novel method for onset detection in musical signals. It improves over previous energy-based and phase-based approaches by combining both types of information in the complex domain. It generates a detection function that is sharp at the position of onsets and smooth everywhere else. Results on a handlabelled data-set show that high detection rates can be achieved at very low error rates. The approach is more robust than its predecessors both theoretically and practically.
Download Extracting automatically the perceived intensity of music titles
We address the issue of extracting automatically high-level musical descriptors out of their raw audio signal. This work focuses on the extraction of the perceived intensity of music titles, that evaluates how energic the music is perceived by listeners. We present here first the perceptive tests that we have conducted, in order to evaluate the relevance and the universality of the perceived intensity descriptor. Then we present several methods used to extract relevant features used to build automatic intensity extractors: usual Mpeg7 low level features, empirical method, and features automatically found using our Extractor Discovery System (EDS), and compare the final performances of their extractors.
Download Perceptually motivated parametric representation for harmonic sounds for data compression purposes
Download Analysis-assisted sound processing with audiosculp
Digital audio effects using phase vocoder techniques are currently in widespread use. However, their interfaces often hide vital parameters from the user. This fact, and the generally limited ways in which sound designing and compositional tools can represent sounds and their spectral content, complicates the effective use of the full potential of modern effect processing algorithms. This article talks about ways in which to use analysis to obtain better processing results with phase vocoder based effects, and how these techniques are implemented in IRCAM's AudioSculpt application. Also discussed are the advantages of using the SDIF format to exchange analysis data between various software components, which facilitates the integration of new analysis and processing algorithms.
Download Interpretable timbre synthesis using variational autoencoders regularized on timbre descriptors
Controllable timbre synthesis has been a subject of research for several decades, and deep neural networks have been the most successful in this area. Deep generative models such as Variational Autoencoders (VAEs) have the ability to generate a high-level representation of audio while providing a structured latent space. Despite their advantages, the interpretability of these latent spaces in terms of human perception is often limited. To address this limitation and enhance the control over timbre generation, we propose a regularized VAE-based latent space that incorporates timbre descriptors. Moreover, we suggest a more concise representation of sound by utilizing its harmonic content, in order to minimize the dimensionality of the latent space.
Download Downmix compatible conversion from mono to stereo in time- and frequency-domain
Even in a time of surround and 3D sound, many tracks and recordings are still only available in mono or it is not feasible to record a source with multiple microphones for several reasons. In these cases, a pseudo stereo conversion of mono signals can be a useful preprocessing step and/or an enhancing audio effect. The conversion proposed in this paper is designed to deliver a neutral sounding stereo image by avoiding timbral coloration or reverberation. Additionally, the resulting stereo signal is downmix-compatible and allows to revert to the original mono signal by a simple summation of the left and right channels. Several configuration parameters are shown to control the stereo panorama. The algorithm can be implemented in time-domain or also in the frequency-domain with additional features, like center focusing.
Download Signal-Matched Power-Complementary Cross-Fading and Dry-Wet Mixing
The blending of audio signals, called cross-fading, is a very common task in audio signal processing. Therefore, digital audio workstations offer several fading curves to select from. The choice of the fading curve typically depends on the signal characteristics and is supposed to result in a mixed signal featuring power and loudness close to the input signals. This work derives a correlationbased design of the fading curves to achieve exact power consistency to avoid audible fluctuations of the signal’s loudness. This principle is extended to the problem of mixing original signals with effect-processed signals using the dry-wet balance. Weighting coefficients for dry and wet signals are derived which realize the desired dry-wet balance but maintain the signal power.
Download Metrix: A Musical Data Definition Language and Data Structure for a Spectral Modeling Based Synthesizer
Since the MIDI 1.0 specification [1], well over 15 years ago, many have been the attempts to give a solution to all the limitations that soon became clear. None of these have had a happy ending, mainly due to commercial interests and as a result, when trying to find an appropriate synthesis control user interface, we had not many choices but the use of MIDI. That’s the reason why the idea of defining a new user interface aroused. In this article, the main components of this interface will be discussed, paying special attention to the advantages and new features it reports to the enduser.
Download Multiband Waveshaping Distortion
A new architecture for musical distortion is proposed. Based on WaveShaping as the distortion generation element, a multiband front-end is used in order to extract simple (ideally monotonal) non-full band signals. Distortion pattern can be adjusted per band, with the benefit that intermodulation distortion is kept low, balancing the end result towards ‘harmonic’ distortion rather than ‘metallic’ or ‘ringing’. Many parameters can be adjusted by the end user in a musically meaningful domain, allowing the creation of rich, detailed and highly personal sounding distortions to be imposed over real world signals.