For many decades, techniques from filter design and system identification have been used to estimate physical parameters of dynamic systems for purposes of simulation and/or control. In the current era, such methods are being extended to employ neural networks. As a result, previously intractable inverse problems are being solved amazingly well by means of local minima in extremely high-dimensional nonlinear optimizations. This presentation will review some highlights of this research and summarize our own efforts in this area to date. Latest results and references can be found on the CCRMA JOS Home Page: https://ccrma.stanford.edu/∼jos/#synth-matching.

The vibration behavior of bowed metallic plates has been understood since Chladni’s pioneering work, but plates made of highly damped, non-metallic materials (paper, cardboard, plastic) reveal additional, less studied effects. When the edge of such a plate is bowed at an oblique angle, the heard pitch suddenly shifts at specific points along the bow stroke. The higher the plate’s fundamental frequency, the more often these jumps occur. Our measurements show that the phenomenon hinges on strong coupling between the bow hair and the plate. Whereas classic Chladni patterns are governed mainly by plate dynamics, the “pitch-jump” effect arises from repeated locking and freeing of the plate to three families of bow-hair modes: 1. modes spanning the entire bow-hair length, 2. ascending modes in the bow-hair segment ahead of the contact point, and 3. descending modes in the segment behind it. A graphical model is introduced that tracks and predicts these mode-locking cycles. Unlocking happens each time the locked mode’s node reaches the contact point, predicting both the positions and the growing density of pitch jumps observed in experiments.

Modal testing is a commonly used method to measure the transfer function or frequency response function of musical instruments or their components.Various excitation and measurement tools are used, and the recorded signals are analyzed to estimate the modal frequencies, damping ratios, and mode shapes of the object.These modal parameters are used to compare musical instruments, study changes to their geometry and materials, create synthesis models, and verify finite element and other simulation models.An object is typically excited with an impact hammer or shaker, while the resulting vibrations are measured with a microphone, accelerometer, or laser Doppler vibrometer.However, musical instrument builders don't typically have access to expensive measurement equipment, so more ad-hoc methods may be used.Multiple methods exist to extract modal parameters from the measured transfer function, each with its own strengths and weaknesses.This study compares commonly used modal extraction methods when applied to measurements of stringed instruments made with a wide spectrum of excitation and measurement sensors.The methods are evaluated based on generated modal data, and then tested with measurements of musical instruments and other objects.

Administration of antibiotics before incision ("surgical antimicrobial prophylaxis") is a critical infection prevention strategy in cardiac surgery. Extending doses of prophylaxis into the postoperative period is common practice in cardiac surgery; however, the benefit has not been clearly established and may drive emergence of antimicrobial resistance and lead to patient harm. Recent World Health Organization guidelines have recommended that prophylaxis be limited to the intraoperative period only for all surgical procedures, but potential benefits with extending prophylaxis postoperatively in cardiac surgery were noted. The efficacy, safety, and microbiome impact of differing durations of surgical antimicrobial prophylaxis in cardiac surgery have not been established.

Ce texte présente mon utilisation du langage Faust pour l'enseignement. J’utilise Faust pour des démonstrations de traitement du signal audio. Je trouve que le moyen le plus efficace d’apprendre la programmation est de « coder en direct » en classe avec l'aide des étudiants. Un langage de haut niveau comme Faust est excellent pour cela, en raison de la quantité réduite et pourtant très lisible du code nécessaire pour avoir des résultats efficaces. J’utilise les « salles de classe inversées » avec des cours enregistrés à l’avance, ce qui laisse plus de temps pour des activités interactives en classe. Une troisième utilisation de Faust est la présentation d’implémentations de référence avec les différentes bibliothèques de Faust. Ces ressources peuvent faire gagner du temps aux étudiants lorsqu'ils doivent mettre en œuvre quelque chose qu'ils connaissent mais qu'ils n'ont pas encore codé. Le fait que Faust compile en C++ et soit exportable immédiatement vers une grande variété de formats de plugin standard le rend très utile comme point de départ pour de nombreux contextes de développement.

Laser Doppler vibrometers (LDV) are used for non-contact vibration measurements of various structures and are frequently used for stringed instrument measurements. Single-point LDVs can be used with the roving hammer or LDV method for mode shape measurements, but this is time-consuming and requires constant attention. Scanning LDVs exist but are expensive and often out of reach of musical acoustics researchers. An inexpensive apparatus to modify a common single-point LDV such that it can perform automated scanning measurements is presented. The augmentation consists of a mirror galvanometer, impact hammer controller, and 3D printed mounting hardware. The scanning system is controlled by a microprocessor and can be easily automated. The total cost of the system, excluding the LDV and impact hammer, is under two hundred dollars. Measurements of guitars are presented to validate the scanning system and discuss any shortcomings.

The modern pipe organ can have as many as 10 000 pipes, each of which is a simple sound generator. In order to digitally synthesize a pipe organ, each pipe requires its own digital waveguide model due to its individual geometry and state. To complicate the matter, pipe geometry can vary greatly across different pipe organ ranks, pipes can be reed-driven or air-jet-driven, and it is possible for a single key to sound many ranks at once. As a result, developing an acoustic model of even a small pipe organ is intensive both in the modeling and in the computational demands for real-time synthesis. For modeling the pipes, we propose a scalable lossy digital waveguide framework whose parameters can be changed based on a pipe's geometric model. We utilize the Faust functional programming language to produce high-performance digital signal processing code.

Digital filters have been used for decades to simulate losses and dispersion in real-time vibrating-string models for musical sound synthesis. String-loop filters in digital-waveguide models have generally been very low-order due to CPU limitations on typical available devices. In recent years, CPU speeds have increased so dramatically that very realistic string models can easily run in real time. This presentation revisits filter design for real-time musical string modeling.

A method is proposed for simulating the frequency-dependent directivity of sound sources for use in interactive geometric acoustic rendering applications. The method is based on using measurements to design a state-space filter allowing the interactive simulation of a time-varying number of radiated sound wavefronts, each towards a time-varying direction. With applicability in sound synthesis or auralization within virtual environments where sound sources dynamically change position and orientation, techniques are proposed for modeling and simulating directivity profiles on perceptually motivated warped frequency axes, along with alternatives for representing directivity on a per-vibration-mode basis or by reduced-order efficient representations. We demonstrate the method by using experimental acoustic data to simulate the directivity of a violin body and a clarinet air column.

A signal processing method to impart the response of an acoustic string instrument to an electric instrument that includes frequency-dependent string decay alterations is proposed. This type of processing is relevant when trying to make a less resonant instrument, such as an electric guitar, sound similar to a more resonant instrument, such as acoustic guitar. Unlike previous methods which typically only perform equalization, our method includes detailed physics-based string damping changes by using a time-varying filter which adds frequency-dependent exponential damping. Efficient digital filters are fit to bridge admittance measurements of an acoustic instrument and used to create equalization filters as well as damping correction filters. The damping correction filters are designed to work in real-time as they are triggered by onset and pitch detection of the signal measured through an under-saddle pickup to determine the intensity of the damping. A test case is presented in which an electric guitar is processed to model a measured acoustic guitar.

The Reflections series takes a look back on historical articles from The Journal of the Acoustical Society of America that have had a significant impact on the science and practice of acoustics.

A method is presented for simulating the free-field, frequency-dependent directivity of linear sound sources for use in real-time within geometric acoustic environments. The method, which is applied to modeling the directivity of a violin body and a clarinet air column from experimental acoustic data in this study, is based on using minimum-phase measurements to design a state-space filter, allowing the interactive simulation of a time-varying number of radiated sound wavefronts, each toward a time-varying direction. With applicability in sound synthesis and/or auralization within virtual environments, where sound sources change position and orientation dynamically, techniques are proposed for modeling and simulating directivity profiles on perceptual frequency axes with alternatives for representing directivity on a per-vibration-mode basis while incorporating relative phase terms or by reduced-order efficient representations comprising separate components for the signature resonant structure and the associated directivity on an adjustable frequency resolution.

Abstract Introduction Mentoring schemes have been established to mitigate these concerns and improve interest in surgery among medical students; however, there is a paucity of research on whether conventional term-time (CTT) mentoring or more intensive out-of-term (IOT) mentoring is more meaningful in promoting surgical careers. Method Most surgical societies in the UK offer CTT schemes (ca. one-to-three mentoring during the academic year) to their members. In addition to this, a student-led surgical society piloted an IOT scheme (one-to-one mentoring during the summer break). Students completed a ten-point questionnaire at the end of the academic year evaluating the scheme(s) they participated in. Result Of the 35 students that completed a questionnaire, 74% participated in either CTT or IOT (1:1), and 26% participated in both. Students that participated in one scheme considered that scheme to be superior in increasing interest in surgery (CTT 69%; IOT 79%), while students that participated in both favoured IOT (78%). With regards to increasing preparedness for surgical careers, IOT was universally preferred. According to previously determined levels of engagement, maximum engagement was achieved by 61.5% of students in a single week of IOT, compared to only 7.7% of students after a full academic year of CTT. Conclusion IOT outperforms CTT with regards to boosting interest in surgery and preparedness for surgical careers. This suggests that introduction of IOT schemes nationally may be a valuable advance in the future of surgical mentoring. Take-home message Adopting an intensive out-of-term mentoring scheme has shown a positive outcome in terms of both increasing preparedness for, and interest in, surgical careers. This suggests implementing such time intensive schemes nationally could be a sensible future-step in surgical mentoring.

In this paper, we present a framework for contrastive learning for audio representations, in a self supervised frame work without access to any ground truth labels. The core idea in self supervised contrastive learning is to map an audio signal and its various augmented versions (representative of salient aspects of audio like pitch, timbre etc.) to a space where they are close together, and are separated from other different signals. In addition we also explore generative models based on state of the art transformer based architectures for learning latent spaces for audio signals, without access to any labels. Here, we map audio signals on a smaller scale to discrete dictionary elements and train transformers to predict the next dictionary element. We only use data as a method of supervision, bypassing the need of labels needed to act as a supervision for training the deep neural networks. We then use a linear classifier head in order to evaluate the performance of our models, for both self supervised contrastive and generative transformer based representations that are learned. Our system achieves considerable performance, compared to a fully supervised method, with access to ground truth labels to train the neural network model. These representations, with avail-ability of large scale audio data show promise in various tasks for audio understanding tasks

A real-time method of string instrument acoustic transfer which includes damping is proposed. Acoustic transfer of string instruments is relevant when trying to make a non-resonant instrument, such as an electric guitar, sound more similar to an acoustic guitar. Unlike previous acoustic transfer methods which only perform equalization, this method includes damping changes by using a time-varying filter which adds frequency-dependent exponential damping. Efficient digital filters are fit to bridge admittance measurements of an acoustic guitar and used to create equalization filters as well as damping correction filters. The damping correction filters work in real-time as they are triggered by onset and pitch detection of the signal measured through an under saddle pickup to determine the intensity of the damping.

This paper proposes a real-time, sample-by-sample pitch tracker for monophonic audio signals using the Extended Kalman Filter in the complex domain (Extended Complex Kalman Filter). It improves upon the algorithm proposed by the same authors in a previous paper [1] by fixing the issue of slow tracking of rapid note changes. It does so by detecting harmonic change in the signal and resetting the filter whenever a significant harmonic change is detected. Along with the fundamental frequency, the ECKF also tracks the amplitude envelope and instantaneous phase of the input audio signal. The pitch tracker is ideal for detecting ornaments in solo instrument music—such as slides and vibratos. The improved algorithm is tested to track pitch of bowed string (double-bass), plucked string (guitar), and vocal singing samples.