Bulk download - click topic to download Zip archive of all papers related to that topic: Acoustics & Speech Audio DSP Capturing HRTFs Mixing Mixing & Education Music AI Musical Acoustics Neural Networks Room Acoustics Soundfield & CTC Soundfields Spatial Audio Spatial Perception Transducers VR/AR VR/Devices/Synthesis
In this paper, differences in the static transfer characteristics of the pre-amplifier of a guitar amplifier were quantitatively evaluated for two types of triode vacuum tubes commonly used by guitarists, the 5751 (amplification factor 70) and the ECC83S (amplification factor 100). First, representative devices were found for each of the two types of vacuum tubes using the parameters of a high-precision physical model, mounted on the pre-amplifier and measurements were performed. Subsequently, modelling by polynomial approximation, a method that provides an intuitive understanding of the relationship between the asymmetric non-linear character of the static transfer characteristics depending on the input signal strength and the even- and odd-order harmonic distortion in the frequency spectrum, and its evaluation were carried out. As a result, it was found that the difference in amplification factor makes a clear difference in the transition from “Clean” to “Overdrive”.
Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member and would like to subscribe to the E-Library then Join the AES!
This paper costs $33 for non-members and is free for AES members and E-Library subscribers.
Start a discussion about this Musical Acoustics!
This paper is a proof of concept application of deep learning technologies towards honeybee colony monitoring. A goal was set to determine internal beehive temperature only through analysis of sound signals produced by the hive. Such a goal was not attempted before. Signals were acquired using an experimental monitoring station, which gathered data from both inside and outside the beehive, as well as recorded temperature inside the beehive. Features extracted from those signals were mel frequency cepstral coefficients and power spectral density. A deep learning convolutional network was employed in the analysis. All tested methods achieved satisfactory results and allowed sufficiently correct prediction of temperatures inside the beehive based on signals recorded by both an internal and an external microphone. Differences of results obtained using external and internal measurements were similar. This proof of concept serves as an indication of future research possibilities concerning automated acoustic monitoring of honeybee families. Such possibilities lie mainly within honeybee health monitoring to which goal this paper’s findings may be of use.
Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member and would like to subscribe to the E-Library then Join the AES!
This paper costs $33 for non-members and is free for AES members and E-Library subscribers.
Start a discussion about this Neural Networks!
Binaural Sound Source Localisation is increasingly being achieved by means of the Convolutional Neural Network (CNN). These networks take in a Time-Frequency representation of audio as an input, and use this to estimate the direction of arrival of a sound. In previous works, different Time-Frequency representations have been used, but never only using solely magnitude spectra, leading to a lack of understanding in the importance of this in full azimuthal binaural sound source localisation. This work aims to address that gap by testing the performance of a CNN trained and tested on four different Time-Frequency representations: Mel-Spectrogram, Gammatonegram, Mel-Frequency Cepstrum, and Gammatone-Frequency Cepstrum. From this test, it was found that Spectrograms are suitable for the task of full azimuthal binaural sound source localisation.
Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member and would like to subscribe to the E-Library then Join the AES!
This paper costs $33 for non-members and is free for AES members and E-Library subscribers.
Start a discussion about this Spatial Audio!
The study presents the project, method of construction, and results of research and analysis of a new type of trumpet mute in relation to its effect on the trumpet sound. The aim was to achieve a new and interesting sound in relation to the other silencers. The main feature of the silencer is its specific construction using alternating modules connected by a tension spring. The principle of the silencer's operation is its complex internal geometry of the modules, forming multiple maze-like systems that modify the sound propagation path. The prototype of the trumpet mute was made by using 3D printing technology based on a previously designed CAD model. The stress analysis was carried out using the finite element method in order to verify the strength of the silencer elements most exposed to damage. The influence of the silencer on the trumpet sound was calculated using the finite element method by analyzing such parameters as directivity characteristics and the sound spectrum. Acoustic measurements were performed in an anechoic chamber on the printed prototype. The influence of the silencer on the performer's intonation was also analyzed. Numerical calculations as well as the acoustic measurements showed a distinct influence of the silencer on the trumpet sound by changing its spectrum, directivity characteristics and by introducing attenuation. In addition to the above analyses, psychoacoustic tests of timbre perception were carried out on a group of random subjects. They revealed significant differences in the perception of trumpet timbre before and after the use of the silencer as well as differences for a trumpet with the silencer or different configurations of the modules.
Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member and would like to subscribe to the E-Library then Join the AES!
This paper costs $33 for non-members and is free for AES members and E-Library subscribers.
Start a discussion about this Musical Acoustics!
In this paper, we propose a rating protocol for evaluating smartphone audio zoom systems through objective and perceptual testing. Audio zoom is a newly developed function that helps isolate a sound source from its surroundings in accordance with the smartphone camera’s focal point and zoom level when recording videos with the camera app. The most important criterion for evaluating a good performance both objectively and perceptually is the device’s ability to focus mainly on the target sound. We also consider and discuss other audio quality criteria; and finally, we conclude by comparing test results and suggesting possible improvements to smartphone audio zoom systems.
Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member and would like to subscribe to the E-Library then Join the AES!
This paper costs $33 for non-members and is free for AES members and E-Library subscribers.
Start a discussion about this VR/Devices/Synthesis!
Surround systems have gained popularity in home entertainment despite the fact that most of the cinematic content is delivered in two-channel stereo format. Although there are several upmixing options, it has proven challenging to deliver an upmixed signal that approximates the original directionality and timbre intended by the mixing artist. The aim of this work is to design a two-to-five channels upmixer using a novel upmixing strategy combining voice extraction and primary-ambience decomposition. Results from a modified-MUSHRA test show that our proposed upmixer outperforms established alternatives for cinematic upmixing in perceived spatial and timbral quality.
Open
Access
Download Now (886 KB)
This paper is Open Access which means you can download it for free.
Start a discussion about this Neural Networks!
Many sound recording methods have been proposed for recording 3D audio. Typically, these methods require at least as many microphones as the number of playback channels. In this study, the authors propose an upmixing method for using fewer microphones. Two methods of upmixing are described, one referring to the MS (mid-side) method and the other generating coherent and diffuse sound components. Using the aforementioned up-mixing technique, the 3-AFC method was used to compare the original 20-channel sound with material converted from 5, 7, and 9 channels to 20 channels. The listening test results showed that the proposed method was perceived as “slightly different” compared to the original 20-channel material unless the material contains a great deal of direct sound in the top layers. Furthermore, from the comments received on the comparison with the original 20 channels, converting a 9-channel or 7-channel signal to 20 channels using the proposed method will result in a higher sense of spaciousness and envelopment.
Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member and would like to subscribe to the E-Library then Join the AES!
This paper costs $33 for non-members and is free for AES members and E-Library subscribers.
Start a discussion about this Spatial Audio!
Head-related transfer functions (HRTFs) represent an essential part in the personalisation of spatial-audio repro-duction over headphones. For frequency regions above 4 kHz, the spectral filtering of the pinna is responsible for the personalisation of an HRTF. In this study, we applied systematic modifications to the pinna geometry of the meshes of ten listeners in order to determine which anatomic regions significantly affect HRTFs on an acoustic level and by means of localisation errors. HRTFs calculated from the modified geometries showed significant spectral differences but only negligible differences in localisation errors predicted by an auditory model for median-plane sound localisation. For example, the backside of the pinna affected the HRTFs in the spectral domain, but did not contribute to localisation errors in the perceptual domain. Our results have implications on future geometry acquisition methods, showing which parts of a pinna do not necessarily need to be captured accurately in order to yield plausible personalised HRTFs by means of localisation errors.
Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member and would like to subscribe to the E-Library then Join the AES!
This paper costs $33 for non-members and is free for AES members and E-Library subscribers.
Start a discussion about this Spatial Audio!
Surface nearfield source (SNS) approach for sound reproduction which is categorized between headphones and loudspeakers was introduced in [1]. The SNS can be embedded for example in the headrest as a personal sound system, providing also a natural audio-tactile augmentation to the listening experience. This paper focuses on the implementation of SNS speaker solution and how it relates to conventional flat panel speakers and structural vi-bration exciter solutions. Operation principle of panel type speakers is discussed, and low frequency vibration behavior of SNS solution integrated into a seat is illustrated using a lumped parameter model.
Open
Access
Download Now (801 KB)
This paper is Open Access which means you can download it for free.
Start a discussion about this Transducers!
This paper examines the suitability of a recurrent neural network, specifically the long/short term memory (LSTM) cell, for black box modelling of electrodynamic loudspeakers. The goal is to develop a versatile and generic nonlinear model that can be applied in industrial settings, such as distortion cancellation and excursion or power limiters. The presented model has the form of a discrete-time single input multiple output (SIMO) system that takes in a digital audio signal and produces membrane displacement and voice coil current as outputs. The training and validation signals used in the model are described, and data from a two-inch broadband loudspeaker driver is used to train the model. The trained LSTM-based model is then compared to a classical state-space model containing the standard displacement-related nonlinearities of force factor Bl(x), inductance L(x) and compliance Cms(x). The parameters of the state-space model were identified using an industry standard method applied to the same two-inch driver. Results show that the LSTM model outperforms the nonlinear state-space model in both time and frequency domains, although it requires longer training time and has a larger model size. A more detailed model comparison follows, and the results are discussed.
Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member and would like to subscribe to the E-Library then Join the AES!
This paper costs $33 for non-members and is free for AES members and E-Library subscribers.
Start a discussion about this Transducers!
