Download Modal Parameter Tracking for Shape-Changing Geometric Objects For interactive sound synthesis, we would like to change the shape of a finite element model of an instrument and rapidly hear how the sound changes. Using modal synthesis methods, we would need to compute a new modal decomposition with each change in the geometry, making the analysis too slow for interactive use. However, by using modes computed for one geometry to estimate the frequencies for nearby geometries, we can hear much more quickly how changing the instrument shape changes the sound. In this paper, we describe how to estimate resonant frequencies of an instrument by combining information about the modes of two similar instruments. We also describe the balance between computational speed and accuracy of the computed resonances.
Download Musical Signal Analysis Using Fractional-Delay Inverse Comb Filters A novel filter configuration for the analysis of harmonic musical signals is proposed. The method is based on inverse comb filtering that allows for the extraction of selected harmonic components or the background noise component between the harmonic spectral components. A highly accurate delay required in the inverse comb filter is implemented with a high-order allpass filter. The paper shows that the filter is easy to design, efficient to implement, and it enables accurate low-level feature analysis of musical tones. We describe several case studies to demonstrate the effectiveness of the proposed approach: isolating a single partial from a synthetic signal, analyzing the even-to-odd ratio of harmonics in a clarinet tone, and extracting the residual from a bowed string tone.
Download Real-Time and Efficient Algorithms for Frequency Warping Based on Local Approximations of Warping Operators Frequency warping is a modifier that acts on sound signals by remapping the frequency axis. Thus, the spectral content of the original sound is displaced to other frequencies. At the same time, the phase relationship among the signal components is altered, nonlinearly with respect to frequency. While this effect is interesting and has several applications, including in the synthesis by physical models, its use has been so far limited by the lack of an accurate and flexible real-time algorithm. In this paper we present methods for frequency warping that are based on local approximations of the warping operators and allow for real-time implementation. Filter bank structures are derived that allow for efficient realization of the approximate technique. An analysis of the error is also presented, which shows that both numerical and perceptual errors are within acceptable limits. Furthermore, the approximate implementation allows for a larger variety of warping maps than that achieved by the classical (non-causal) first-order allpass cascade implementation.
Download Short-Time Wavelet Analysis of Analytic Residuals for Real-Time Spectral Modelling This paper describes an approach to using compactly supported spline wavelets to model the residual signal in a real-time (frameby-frame) spectral modelling system. The outputs of the model are time-varying parameters (gain, centre frequency and bandwidth) for filters which can be used in a subtractive resynthesis system.
Download A Complex Envelope Sinusoidal Model for Audio Coding A modification to the hybrid sinusoidal model is proposed for the purpose of high-quality audio coding. In our proposal the amplitude envelope of each harmonic partial is modeled by a narrowband complex signal. Such representation incorporates most of the signal energy associated with sinusoidal components, including that related to frequency estimation and quantization errors. It also takes into account the natural width of each spectral line. The advantages of such model extension are a more straightforward and robust representation of the deterministic component and a clean stochastic residual without ghost sinusoids. The reconstructed signal is virtually free from harmonic artifacts and more natural sounding. We propose to encode the complex envelopes by the means of MCLT transform coefficients with coefficient interleave across partials within an MPEG-like coding scheme. We show some experimental results with high compression efficiency achieved.
Download Warped Linear Prediction for Improved Perceptual Quality in the SCELP Low Delay Audio Codec The SCELP (Spherical Code Excited Linear Prediction) audio codec, which has recently been proposed for low delay audio coding [5], is based on linear prediction (LP). It applies closed-loop vector quantization employing a spherical code which is based on the Apple Peeling code construction rule. Frequency warped signal processing is known to be beneficial especially in the context of wideband audio coding based on warped linear prediction (WLP). In this contribution, WLP is incorporated into the SCELP low delay audio codec. The overall audio quality of the resulting W-SCELP codec benefits from an improved perceptual masking of the quantization noise. Compared with existing standardized audio codecs with an algorithmic delay below 10 ms, the W-SCELP codec at a data rate of 48 kbit/sec outperforms the ITU-T G.722 codec at a data rate of 56 kbit/sec in terms of the achievable audio quality.
Download Transient Encoding of Audio Signals Using Dyadic Approximations In this paper, we present a frame based approach for transient detection and encoding of audio signals. The transient detection procedure, as presented here, uses linear prediction within a signal frame followed by an envelope estimation to build an adaptive threshold. Detected transients will automatically be separated and the gaps left by the removed transient are filled with samples from forward and backward extrapolation. To encode detected transients, dyadic approximation approaches are discussed. Results of the application to different audio signals are also presented.
Download Adjustable Boundary Conditions for Multidimensional Transfer Function Models Block based physical modeling requires to provide a library of modeling blocks for standard components of real or virtual musical instruments. Complex synthesis models are built by connecting standard components in a physically meaningful way. These connections are investigated for modeling a resonating structure as a distributed parameter system. The dependence of a resonator’s spectral structure on the termination of its ports is analyzed. It is shown that the boundary conditions of a distributed parameter system can be adjusted by proper termination only. Examples show the corresponding variation of the resonator’s spectral structure in response to variations of the external termination.
Download Self-Sustained Vibrating Structures Physical Modelling by Means of Mass-Interaction Networks GENESIS is a sound synthesis and musical creation environment based on the mass-interaction CORDIS-ANIMA physical modelling formalism. It has got the noteworthy property that it allows to work both on sound itself and on musical composition in a single coherent environment. In this paper we present the first results of a study that is carried out with GENESIS on a particular type of models: self-sustained oscillating structures. By trying to build physical models of real instruments like bowed strings or woodwinds, our aim is to develop and analyse generic tools that can be used for the production of self-sustained oscillations on every mass-interaction network built with GENESIS. But, if the family of the self-sustained oscillating structures is very interesting to create rich timbres, it can also play a new and fundamental role at the level of the temporal macrostructure of the music (that of the gesture and the instrumental performance, as well as the composition). Indeed, it is possible, as we will propose in this paper, to use the relatively complex motion of a bowed macrostructure in a musical composition way, as a musical events generator.
Download Filtering within the Framework of Mass-Interaction Physical Modeling and of Haptic Gestural Interaction A variety of filters have been designed, synthesized and used in the history of electronic and computer music. All the approaches aimed to provide filters fulfilling several specifications such as frequency response, phase response, transient state characteristics like rise time and overshoot, realizably conditions concerning the technology used for the implementation and even economical considerations. One of the most important aspects concerning the filters dedicated to musical applications is the control structure they provide to the musician, who is in charge for the integration of the filtering operation in the compositional process and performance. Designing filters using the mass interaction scheme embedded in the CORDIS-ANIMA formalism (used for sound synthesis and composition by physical modelling) offers a different methodology in the control which is coherent with the philosophy of musical composition by ‘physical thinking’. This article introduces a technique to design filters using the CORDIS-ANIMA simulation language.