Browsing by Author "Välimäki, Vesa, Prof., Aalto University, Department of Signal Processing and Acoustics, Finland"

Now showing 1 - 9 of 9
  • Thumbnail Image
    Item
    Analysis and Synthesis of Directional Reverberation
    (Aalto University, 2021) Alary, Benoit; Politis, Archontis, Dr., Tampere University, Finland; Schlecht, Sebastian J., Prof., Aalto University, Finland; Välimäki, Vesa, Prof., Aalto University, Finland; Signaalinkäsittelyn ja akustiikan laitos; Department of Signal Processing and Acoustics; Audio Signal Processing; Sähkötekniikan korkeakoulu; School of Electrical Engineering; Välimäki, Vesa, Prof., Aalto University, Department of Signal Processing and Acoustics, Finland
    The reproduction of acoustics is one of the key challenges in spatial sound reproduction. In order to reach high levels of realism and immersion, reproduced signals must be processed through a set of filters accounting for real-life phenomena. From the scattering and absorption of sound on walls to its diffraction around objects and our head, the propagation of sound waves in a room creates a complex sound field around a listener. While recreating all of the underlying parts of sound propagation is not yet within our reach for real-time applications, spatial sound techniques aim to minimize the computational cost by focusing the efforts on the most perceptually critical components. In this dissertation, the reproduction of reverberant sound fields is investigated through the development of analysis methods, reverberation algorithms, and decorrelation techniques. A particular emphasis is given to the perception and reproduction of directional characteristics present in late reverberation. Most reverberation algorithms consider that the sound energy is evenly distributed across space in all directions, after an initial period of time, since the diffusion of energy leads to more homogeneous and isotropic sound fields. However, previous work has demonstrated that insufficient diffusion in a room leads to anisotropic, and directional, late reverberation. In this dissertation, a complete framework is proposed for the objective and subjective analysis of directional characteristics as well as a novel delay-network reverberation method capable of producing direction-dependent decay properties. The reverberator is also expanded to offer efficient frequency- and direction-dependent processing. The proposed algorithm contains all the required elements for the auralization of reverberant sound fields, which may be modulated in real-time to support six-degrees-of-freedom sound reproduction. The decorrelation of audio signals, which occurs naturally during the diffusion of energy in a sound field, is another important aspect of sound reproduction. This dissertation considers the use of velvet-noise sequences, a special type of sparse noise signals, as decorrelation filters and offers a method to optimize their characteristics. Velvet noise is also proposed for the reproduction of an existing impulse response using a small set of time- and frequency-dependent information, along with a reverberator using velvet noise to improve the echo density of a delay network reverberator. Overall, the results contained in this dissertation offer new insights into the perceptual ramifications of reverberant sound fields containing directional characteristics and their reproduction. The methods presented bring applications in the context of immersive sound reproduction, such as in virtual and augmented reality.
  • Thumbnail Image
    Item
    Circuit modeling studies related to guitars and audio processing
    (Aalto University, 2013) Dias de Paiva, Rafael; Pakarinen, Jyri, PhD, Dolby Laboratories, Sweden; Signaalinkäsittelyn ja akustiikan laitos; Department of Signal Processing and Acoustics; Laboratory of Acoustics and Audio Signal Processing; Sähkötekniikan korkeakoulu; School of Electrical Engineering; Välimäki, Vesa, Prof., Aalto University, Department of Signal Processing and Acoustics, Finland
    This thesis addresses the use of circuit modeling techniques in audio. Circuit modeling has a wide range of applications in audio, including real-time models of analog electronic audio equipment and the use of physical analogies for understanding and simulating musical instru-ments. Modeling of analog audio equipment is an important topic in audio signal processing. It enables the development of musical software that is capable of simulating rare vintage equip-ment at a low cost. This type of software can be embedded in portable electronic equipment, in mobile phones or tablets, or in computers. This thesis presents novel models of analog audio equipment used with guitars. It presents a nonlinear audio-transformer model which is used for real-time emulation of vacuum-tube guitar amplifiers. This model has shown that some audio transformers have nonlinear effects for input signals with frequencies below 100 Hz. A new wave-digital model for operational amplifiers is proposed, which is used to simulate a wide class of guitar distortion circuits. The same distortion circuits were modeled with a novel method based on nonlinear system identi-fication, which is enhanced using principal component analysis (PCA) for reduced complexity. It was shown that the proposed method reduces the complexity of the polynomial-Hammer-stein model obtained with the swept-sine technique by 66 %. Additionally, electromagnetic pickups were analyzed and modeled, leading to new pickup-mixing and nonlinearity models and to a better understanding on the effects of guitar pickup and cable interaction. This thesis has also presented how to use physical analogies for audio synthesis. Electro-acous-tic analogies were used in order to obtain a model of connected Helmholtz resonators, resulting in the so called Helmholtz resonator tree. This model was implemented using wave-digital filters, which enables musical synthesis using physical descriptors that are intuitive also for non-technical users. This thesis includes contributions for the application of circuit modeling techniques in audio. The audio transformer, electromagnetic pickup, and effect-box modeling developments are important for building real-time systems for audio effects and for preserving the heritage of vintage analog equipment. Finally, the electro-acoustic analogies presented show that circuit modeling can be used for abstract musical synthesis, where a virtual instrument can be excited in different manners yielding interesting timbre variations.
  • Thumbnail Image
    Item
    Dispersive Systems in Musical Audio Signal Processing
    (Aalto University, 2013) Parker, Julian D.; Välimäki, Vesa, Prof., Aalto University, Department of Signal Processing and Acoustics, Finland; Signaalinkäsittelyn ja akustiikan laitos; Department of Signal Processing and Acoustics; Audio Signal Processing Team; Sähkötekniikan korkeakoulu; School of Electrical Engineering; Välimäki, Vesa, Prof., Aalto University, Department of Signal Processing and Acoustics, Finland
    Dispersion is a property seen throughout both nature and the man-made world. By its most basic definition, it simply refers to the spreading out or scattering of some form of wave phenomenon in a medium. In practice, this is usually due to variation in the propagation speed of the wave with respect to frequency or amplitude. Examples of dispersion in the natural world are rainbows, the spreading of water surface waves, and the distinctive sound generated when striking a long metal wire. This thesis explores the presence of dispersion in signal processing systems designed for musical use, and develops a number of new musical digital signal processing systems which are based around the action of dispersion. The primary focus of the thesis is spring reverberation, an early form of artificial reverberation based on exciting vibrations in helical metal springs. Springs are unique in the world of analog musical technology, in that they derive almost all of their recognized sonic character and desirability from the dispersion they induce. The first portion of this thesis examines the behavior of spring reverberators through the lens of mathematical models of spring vibration, and develops some important new results about their behavior. These mathematical models are turned directly into a digital model, via the application of finite difference methods. A similar system, that of the larger 'Slinky' spring, is examined and modeled via the same framework. The second part of this thesis examines an alternative method for emulating spring reverberation, based heavily on the use of the dispersive allpass filter. The result is an efficient and high-quality parametric emulation of the spring reverberator. This model is further extended in several novel ways to improve its efficiency, including via application of multi-rate techniques. The reverberator is further improved by the proposal of a new structure that uses sparse-noise convolution to generate diffusion in the repeating echoes. Finally, a new algorithm is developed that uses the dispersion of golden-ratio allpass filters as a method for transparently reducing the peak amplitude of musical signals so that their loudness can be maximized. The novel algorithms developed in this thesis are intended for use in real-time music production or processing applications. Hence, computational efficiency and parametric control are central considerations. These algorithms could be implemented for use in a computer environment, for mobile devices, or for embedded DSP systems.
  • Thumbnail Image
    Item
    Equalization Techniques for Headphone Listening
    (Aalto University, 2014) Rämö, Jussi; Välimäki, Vesa, Prof., Aalto University, Department of Signal Processing and Acoustics, Finland; Signaalinkäsittelyn ja akustiikan laitos; Department of Signal Processing and Acoustics; Sähkötekniikan korkeakoulu; School of Electrical Engineering; Välimäki, Vesa, Prof., Aalto University, Department of Signal Processing and Acoustics, Finland
    The popularity of headphones has increased rapidly along with digital music and mobile phones. The environment in which headphones are used has also changed quite dramatically from silent to noisy, since people are increasingly using their headphones while commuting and traveling. Ambient noise affects the quality of the perceived music as well as compels people to listen to the music with higher volume levels. This dissertation explores headphone listening in the presence of ambient sounds. The ambient sounds can be either noise or informative sounds, such as speech. The first portion of this work addresses the first case, where the ambient sounds are undesirable noise that deteriorates the headphone listening experience. The second portion tackles the latter case, in which the ambient sounds are actually informative sounds that the user wants to hear while wearing headphones, such as in an augmented reality system. Regardless of the nature of the ambient sounds, the listening experience can be enhanced with the help of equalization. This work presents a virtual listening test environment for evaluating headphones in the presence of ambient noise. The simulation of headphones is implemented using digital filters, which enables arbitrary music and noise test signals in the listening test. The disturbing effect of ambient noise is examined with the help of a simulator utilizing an auditory masking model to simulate the timbre changes in music. Another study utilizes the same principles and introduces an adaptive equalizer for mitigation of the masking phenomenon. This psycho- acoustic audio processing system was shown to retain reasonably low sound pressure levels while boosting the music, which is highly important from the viewpoint of hearing protection. Furthermore, two novel hear-through systems are proposed, the first of which is a digital augmented reality headset substituting and improving a previous analog system. The second system is intended to be worn during loud concerts as a user-controllable hearing protector and mixer. The main problem in both of the systems is the risk of a comb filtering effect, which can deteriorate the sound quality. However, it is shown that the comb filtering effect is not detrimental due to the passive isolation of an in-ear headset. Finally, an optimization algorithm for high-order graphic equalizer is developed, which optimizes the order of adjacent band filters to reduce ripple around the filter transition bands. Furthermore, a novel high-precision graphic equalizer is introduced based on parallel second- order sections. The novel equalization techniques are intended for use not only in headphone applications, but also in wide range of other audio signal processing applications, which require highly selective equalizers.
  • Thumbnail Image
    Item
    Equalizer Design for Sound Reproduction
    (Aalto University, 2020) Liski, Juho; Rämö, Jussi, Dr., Aalto University, Finland; Välimäki, Vesa, Prof., Aalto University, Finland; Signaalinkäsittelyn ja akustiikan laitos; Department of Signal Processing and Acoustics; Aalto Acoustics Lab; Sähkötekniikan korkeakoulu; School of Electrical Engineering; Välimäki, Vesa, Prof., Aalto University, Department of Signal Processing and Acoustics, Finland
    Digital graphic equalizers are widely used in various applications, such as music production and reproduction, to control the magnitude responses of signals. They are straightforward to operate, with the gain at each band being the only controllable parameter. Different band divisions and different equalizer structures are used based on the requirements of the application or the available hardware. However, the accurate design of graphic equalizers is not straightforward due to band leakage, i.e., the interaction of the neighboring band filters leading to approximation errors. This is especially problematic in applications utilizing automatic equalizer updates, such as the adaptive equalization of headphones or loudspeakers based on ambient noise. In these applications, the target curve must be realized accurately and efficiently. This thesis focuses on accurate graphic-equalizer design and its applications. The first part discusses different equalizer structures and proposes a novel parallel graphic equalizer with a delayed IIR part. Almost all discussed designs achieve an accuracy of ±1 dB with the help of band filters whose responses are exactly controlled at three points. A least-squares method jointly optimizes the filter gains and accounts for the band leakage. It is shown that the number of design points should be double the number of bands, the accuracy is increased with band filters having a symmetric shape up to the Nyquist frequency, and the design parameters must be optimized for each design separately. The second part of the thesis discusses equalizer applications related to headphones. The proposed algorithms apply adaptive filters to hear-through processing and to individualization of the headphone response. The former algorithm estimates the isolation of the headphones and controls the equalizer in order to achieve acoustic transparency despite changes in the headphone fit. The latter example applies a similar adaptive filter to estimate the headphone magnitude response, which is then equalized to a selected target. Measurements indicate that both the magnitude-response estimation and equalization work accurately. The final part of the thesis focuses on loudspeaker equalization, and both the magnitude-only and phase equalization are elaborated. Multiple flat-panel loudspeakers are measured, and their sound is improved using a single-point equalizer design. A simulated loudspeaker is group-delay equalized with a frequency-sampled FIR filter, and a limited-band equalizer is shown to be supe- rior to the full-band group-delay equalization. The proposed accurate equalizers are suitable for a wide range of audio system equalization applications, both human- and algorithm-operated ones. The different structures enable the equalizers to be run on different hardware, and the low complexity of the equalizers comprising a single second-order section per band makes them especially suitable for mobile implementations.
  • Thumbnail Image
    Item
    Filter-Based Oscillator Algorithms for Virtual Analog Synthesis
    (Aalto University, 2014) Pekonen, Jussi; Signaalinkäsittelyn ja akustiikan laitos; Department of Signal Processing and Acoustics; Sähkötekniikan korkeakoulu; School of Electrical Engineering; Välimäki, Vesa, Prof., Aalto University, Department of Signal Processing and Acoustics, Finland
    This thesis deals with virtual analog synthesis, i.e., the digital modeling of subtractive synthesis principle used in analog synthesizers. In subtractive synthesis, a spectrally rich oscillator signal is modified with a time-varying filter. However, the trivial implementation of the oscillator waveforms typically used in this synthesis method suffers from disturbing aliasing distortion. Filter-based algorithms that produce these waveforms with reduced aliasing are studied in this thesis. An efficient antialiasing oscillator technique expresses the waveform as a bandlimited impulse train or a sum of time-shifted bandlimited step functions. This thesis proposes new polynomial bandlimited function generators and introduces optimized look-up table and polynomial-based functions for these algorithms. A new technique for generating nonlinear-phase bandlimited functions is also presented. In addition to the aforementioned technique, the research focus in oscillator algorithms is on ad-hoc approaches that either post-process the output of the trivial oscillator algorithm or produce signals that look similar to the classical waveforms. Linear post-processing algorithms that suppress aliasing of the waveform generated, in principle, by any oscillator algorithm are introduced in this thesis. Perceptual aspects of the audibility of aliasing are also addressed in this thesis. The results of a listening test that studied the audibility of aliasing distortion in a trivially sampled sawtooth signal are shown. Based on the test results, design criteria for digital oscillator algorithms are obtained and the usability of previously used computational measures for the evaluation of aliasing audibility is analyzed. In addition, modeling of analog synthesizer oscillator outputs is addressed in this thesis. Two separate models for the sawtooth signal generated by the oscillator circuitry of the MiniMoog Voyager analog synthesizer are developed. The first model uses phase distortion to generate sawtooth waveforms that resemble that of the MiniMoog. The second model filters the output of a digital oscillator algorithm with a fundamental frequency dependent post-processing filter. The techniques described in this thesis can be used in the development of alias-free oscillator algorithms for virtual analog synthesis. Also, the output of this oscillator can be processed to sound and look like the respective waveform of any analog synthesizer using the methods proposed here.
  • Thumbnail Image
    Item
    Measurement and Prediction of the Spatial Decay of Speech in Open-Plan Offices
    (Aalto University, 2015) Keränen, Jukka; Hongisto, Valtteri, Adj. Prof., Finnish Institute of Occupational Health, Finland; Signaalinkäsittelyn ja akustiikan laitos; Department of Signal Processing and Acoustics; Sähkötekniikan korkeakoulu; School of Electrical Engineering; Välimäki, Vesa, Prof., Aalto University, Department of Signal Processing and Acoustics, Finland
    This thesis consists of seven publications that presented results of room acoustic research studies. Room acoustic measurement and prediction methods for work spaces are studied. Measurements are made in industrial workrooms, open-plan offices and two significantly different laboratory spaces of Finnish Institute of Occupational Health, Turku. This thesis presents a new room acoustic measurement method for open-plan offices. The method is published in ISO 3382-3:2012 standard. The method is developed from a method originally used in room acoustic measurements of industrial workrooms. An omnidirectional loudspeaker and wide band noise signal are used to measure spatial decay of sound pressure level (SPL) of speech from one workstation to other workstations on a measurement line. SPL of masking sound and speech transmission index (STI) are also measured in the workstations. The measurement results are used to determine single-number quantities: spatial decay rate of the A-weighted SPL of speech, D_2,S, the A-weighted SPL of normal speech in the distance of 4 m from the speaker, L_p,A,S,4m, and distraction distance, r_D. Target values are determined for the single-number quantities. They are based on the distribution of measurement results in Finnish open-plan offices. Simple prediction models are developed for D_2,S and L_p,A,S,4m. The models are based on linear regression analysis of empirical data from 16 different open-plan offices. The effect of room acoustical changes on spatial decay of speech in open-plan offices can be estimated using these simple models. A procedure to calculate STI using the simple models and to determine r_D is described. The predicted single-number quantities are compared to measured quantities in 26 open-plan offices. The effect of room acoustic changes on STI and spatial decay of speech is studied in two series of laboratory experiments. The first study focuses on two adjacent workstations in a test room where all the walls are sound-absorbing. The effects of ceiling, floor and screen absorption, screen height, room height and masking sound level are investigated. The second study examines the effect of ceiling, wall and screen absorption, screen height and masking sound level on the A-weighted SPL of speech, the spatial decay rate of speech and STI in an open-plan office laboratory with 12 workstations. The laboratory experiments provide measured evidence on the effects that typical room acoustic changes cause to SPL of speech and STI in open-plan offices. The effect of ceiling, wall and screen absorption, screen height and SPL of masking sound are studied exhaustively. The results emphasize the importance of all-inclusive design of room acoustics in open-plan offices.
  • Thumbnail Image
    Item
    Room Reverberation Prediction and Synthesis
    (Aalto University, 2022) Prawda, Karolina Anna; Schlecht, Sebastian J., Prof., Aalto University, Finland; Välimäki, Vesa, Prof., Aalto University, Finland; Signaalinkäsittelyn ja akustiikan laitos; Department of Signal Processing and Acoustics; Audio Signal Processing; Sähkötekniikan korkeakoulu; School of Electrical Engineering; Välimäki, Vesa, Prof., Aalto University, Department of Signal Processing and Acoustics, Finland
    In this dissertation, the discussion is centered around the sound energy decay in enclosed spaces. The works starts with the methods to predict the reverberation parameters, followed by the room impulse response measurement procedures, and ends with an analysis of techniques to digitally reproduce the sound decay. The research on the reverberation in physical spaces was initiated when the first formula to calculate room's reverberation time emerged. Since then, finding an accurate and reliable method to predict reverberation has been an important area of acoustic research. This thesis presents a comprehensive comparison of the most commonly used reverberation time formulas, describes their applicability in various scenarios, and discusses their accuracy when compared to results of measurements. The common sources of uncertainty in reverberation time calculations, such as bias introduced by air absorption and error in sound absorption coefficient, are analyzed as well. The thesis shows that decreasing such uncertainties leads to a good prediction accuracy of Sabine and Eyring equations in diverse conditions regarding sound absorption distribution. The measurement of the sound energy decay plays a crucial part in understanding the propagation of sound in physical spaces. Nowadays, numerous techniques to capture room impulse responses are available, each having its advantages and drawbacks. In this dissertation, the majority of commonly used measurement techniques are listed, whereas the exponential swept-sine is described in more detail. This work elaborates on the external factors that may impair the measurements and introduce error to their results, such as stationary and non-stationary noise, as well as time variance. The dissertation introduces Rule of Two, a method of detecting non-stationary disturbances in sweep measurements. It also shows the importance of using median as a robust estimator in non-stationary noise detection. Artificial reverberation is a popular sound effect, used to synthesize sound energy decay for the purpose of audio production. This dissertation offers an insight into artificial reverberation algorithms based on recursive structures. The filter design proposed in this work offers precise control over the decay rate while being efficient enough for real-time implementation. The thesis discusses the role of the delay lines and feedback matrix in achieving high echo density in feedback delay networks. It also shows that four velvet-noise sequences are sufficient to obtain smooth output in interleaved velvet noise reverberator. The thesis shows that the accuracy of reproduction increases the perceptual similarity between measured and synthesised impulse responses. The insights collected in this dissertation offer insights into the intricacies of reverberation prediction, measurement and synthesis. The results allow for reliable estimation of parameters related to sound energy decay, and offer an improvement in the field of artificial reverberation.
  • Thumbnail Image
    Item
    Virtual Analog Modeling of Nonlinear Musical Circuits
    (Aalto University, 2014) D'Angelo, Stefano; Signaalinkäsittelyn ja akustiikan laitos; Department of Signal Processing and Acoustics; Laboratory of Acoustics and Audio Signal Processing; Sähkötekniikan korkeakoulu; School of Electrical Engineering; Välimäki, Vesa, Prof., Aalto University, Department of Signal Processing and Acoustics, Finland
    Recent advances in semiconductor technology eventually allowed for affordable and pragmatic implementations of sound processing algorithms based on physical laws, leading to considerable interest towards research in this area and vast amounts of literature being published in the last two decades. As of today, despite the efforts invested by the academic community and the music technology industry, new or better mathematical and computational tools are called for to efficiently cope with a relatively large subset of the investigated problem domain. This is especially true of those analog devices that inherently need to be studied by lumped nonlinear models. This research is, in this sense, directed towards both general techniques and specific problems. The first part of this thesis presents a generalization of the wave digital filter (WDF) theory to enable interconnections among subnetworks using different polarity and sign conventions. It proposes two new non-energic two-port WDF adaptors, as well as an extension to the definitions of absorbed instantaneous and steady-state pseudopower. This technique eventually removes the need to remodel subcircuits exhibiting asymmetrical behavior. Its correctness is also verified in a case study. Furthermore, a novel, general, and non-iterative delay-free loop implementation method for nonlinear filters is presented that preserves their linear response around a chosen operating point and that requires minimal topology modifications and no transformation of nonlinearities. In the second part of this work, five nonlinear analog devices are analyzed in depth, namely the common-cathode triode stage, two guitar distortion circuits, the Buchla lowpass gate, and a generalized version of the Moog ladder filter. For each of them, new real-time simulators are defined that accurately reproduce their behavior in the digital domain. The first three devices are modeled by means of WDFs with a special emphasis on faithful emulation of their distortion characteristics, while the last two are described by novelly-derived systems in Kirchhoff variables with focus on retaining the linear response of the circuits. The entirety of the proposed algorithms is suitable for real-time execution on computers, mobile electronic devices, and embedded DSP systems.