AES Fall Online Convention

This work introduces an Open Source turntable for the measurement of electro-acoustical devices. The idea is to provide an inexpensive and highly customizable device that can be adjusted according to specific measurement needs. Development of such turntable devices in the past required significant investment. Specific mechanical and motor control design skills were needed, leading to both costly and time-consuming processes. Recent developments in mechatronics and 3D printing allow to design and build a cost-effective solution.

This e-Brief proposes a novel approach of audio watermarking based on Spread Spectrum (SS) involves the psychoacoustic model and deep learning Convolutional Neural Networks (CNN)-autoencoder. Moreover, logistic chaotic maps are employed to enhance the security level of the method. First, a compressed image produced from the CNN-autoencoder is fed to the image encryption stage to yield an encrypted image to be used as the watermark. To apply image encryption, the plain image is, at first 8-bit binary-coded and shuffled by M-sequence. Next, each encoded image is diffused with a different chaotic sequence. Within the embedding phase, the psychoacoustic model is employed to shape the amplitude of the watermark signal which guarantees high inaudibility, whereas a logistic chaotic map is used to determine the positions for watermark embedding in a random manner. This scheme offers an extremely efficient and practical method as it can be used by institutions and companies for embedding their logos or trademarks as a watermark in audio products as the scheme utilizes RGB images. Experimental results show that the transparency and imperceptibility of the proposed algorithm are satisfactory also good image quality even against various attacks. The validity of the proposed audio watermarking method is demonstrated by simulation results.

This paper discovers methods to create vibrato effects with artificial reverberation. Feedback delay networks have been used for many reverb effects. Past work with time varying feedback delay networks has focused primarily on small modulations of the delays and or feedback matrices in order to create a more natural sounding reverb. In this paper, we consider the possibility of using wider modulations of these reverbs for the purposes of sound effect generation. Specifically, amplitude modulation and frequency modulation can be obtained by varying feedback matrices or delay lines respectively. The results showed a convincing vibrato effect with minor artifacts and promise for using FDNs in sound effect generation. Future work will include reducing artifacts and fine tuning control parameters.

While deep learning approaches for beat and downbeat tracking have brought advancements, these approaches continue to rely on hand-crafted, subsampled spectral features as input, restricting the information available to the model. In this work, we propose WaveBeat, an end-to-end approach for joint beat and downbeat tracking operating directly in the time domain. This method forgoes engineered spectral features, and instead produces beat and downbeat predictions directly from the waveform, the first of its kind for this task. Our model utilizes temporal convolutional networks (TCNs) operating on waveforms that achieve a very large receptive field (~30s) at audio sample rates. This is achieved in a memory efficient manner by employing rapidly growing dilation factors, which enable a relatively shallower network architecture. Combined with a straightforward data augmentation strategy, our method outperforms previous state-of-the-art methods on some datasets, while producing comparable results on others.

This paper offers a study of the circuits developed by artist Paul DeMarinis for the touring version of his work Pygmy Gamelan. Each of the six copies of the original circuit, developed June-July 1973, produce a carefully tuned and unique five-tone scale. These are obtained by five resonator circuits which pitch pings produced by a crude antenna fed into clocked bit-shift registers. While this resonator circuit may seem related to classic Bridged-T and Twin-T designs, common in analog drum machines, DeMarinis' work actually presents a unique and previously undocumented variation on those canonical circuits. We present an analysis of his third-order resonator (which we name the Gamelan Resonator), deriving its transfer function, time domain response, poles, and zeros. This model enables us to do two things: first, based on recordings of one of the copies, we can deduce which standard resistor and capacitor values DeMarinis is likely to have used in that specific copy, since DeMarinis' schematic purposefully omits these details to reflect their variability. Second, we can better understand what makes this filter unique. We conclude by outlining future projects which build on the present findings for technical development.

Hearing protection devices (HPDs) are essential for musicians during loud performances to avoid hearing damage, but the standard Noise Reduction Rating (NRR) performance metric for HPDs metric says little about their behavior in a musical setting. One analysis tool being used to evaluate HPDs in the noise exposure research community is kurtosis measured in the ear and the reduction of noise kurtosis through an HPD. A musical signal, especially live music, will often have a high crest factor and kurtosis, so evaluating kurtosis loss will be important for an objective evaluation of musicians' HPDs. In this paper, a background on kurtosis and filters affecting kurtosis is described, as well as a setup for generating high-kurtosis signals and measuring in-ear kurtosis loss through an HPD. Measurement results on a variety of musicians' HPDs show that 83% of devices measured strongly reduce kurtosis, and that the kurtosis loss is likely an independent metric for performance because it is not correlated to the mean or standard deviation of the spectral insertion loss.

Created in conjunction with the Marine Institute at the University of Plymouth, the intention of this project was to use data transmitted by the on-board sensors of the Mayflower Autonomous Ship (MAS), to manipulate specially created pieces of music, based on sea shanties and folk ballads. Technical issues and Covid delays forced a late change, and the project was switched to using data from the university’s weather stations. This paper will illustrate how the music was produced and recorded, and the software configured to make the musical pieces vary in real-time, according to the changing sea conditions, so that the public will be able to view the current conditions and listen to the music evolve in real-time.

Over recent years, numerous attempts were taken to provide efficient methods of directivity representation, either regarding sound sources or head-related transfer functions. Because of the wide variety of programming tools and scripts used by different researchers, the resulting representations are inconvevnient to reproduce and compare with each other, hampering the development of the subject. Within this paper, an objective-oriented method is proposed to deal with this issue. The suggested approach bases on defining classes for different directivity models that share some general properties of directivity functions, allowing for easy comparison between different representations. A basic Matlab toolbox utlizing this method is presented alongside exemplary implementations of directivity models based on spherical and hyperspherical harmonics.

We present a parametric array system realized with a microcontroller and MOSFET drivers. Pulse train signals with fundamental frequency of 40 kHz are generated by the microcontroller. The pulse trains are amplitude modulated by exploiting the switching mechanism of the MOSFETs. The higher-order harmonics are attenuated by the band-pass characteristic of the ultrasonic transducers, emitting only the carrier frequency and the sideband components. The sound beam can be steered by applying phase shifts to the pulse signals, which can be implemented by relatively inexpensive hardware. A new single sideband modulation is also introduced, where the upper sidebands of two double sideband modulation signals are acoustically cancelled. The proposed approaches for beamforming and single sideband modulation are evaluated by anechoic measurements.