Search

The purpose of the article is to characterize the principles of sound recording with microphones in the context of the acoustic spatial features of concert halls, which are an important component in positioning the activities and creative directions of "purism", "individualism" and "realism" in sound engineering. The methodology consists of the use of analytical, historical, and cultural methods, which made it possible to identify and characterize the technological foundations of sound recording using the example of sound engineers. The scientific novelty of the work lies in the fact that for the first time in Ukrainian science the principles of microphone sound recording in the context of acoustic spatial features and creative directions of sound engineering "purism", "individualism" and "realism" were defined and characterized. Conclusions. In the work, the data on the spectral response of the frequency range, the stereophonic effect, musical and timbre balance, and the spatial impression of the acoustics of concert halls were determined. The principles of application of multi-microphone technique in instrumental, orchestral, and rock music are revealed; outlined the creative potential of the directions "purism", "individualism" and "realism". in sound engineering. In terms of current cinematic trends and contemporary popular music culture, we hear and become accustomed to exaggeratedly colorful and rich, often "electronic" sound. Since the listener is the ultimate link in the entire recording industry, it is necessary to recognize landmarks in sound engineering aimed at the tastes of the majority.

As a challenging task of audio processing, automatic music transcription (AMT) attracts increasing attention recently, which aims to convert a raw audio to a symbolic representation. Nowadays, music recordings are usually stereo audio files. Many previous studies simply average the stereo signal to a mono signal when processing data, which sacrifices some useful information. In this paper, we design a stereo feature enhancement (SFE) module based on self-attention mechanism to make full use of stereo information. Moreover, in recent years temporal convolutional network (TCN) has demonstrated great effect on processing temporal data, which overcomes some drawbacks of existing temporal information extraction methods such as HMM, RNN and LSTM. Inspired by this, we propose a temporal convolutional module (TCM) which is suitable to extract temporal context of music. Our proposed network is validated on the MAPS dataset for music transcription, and achieves ideal performance.

This essay is prompted by “surround sound,” the sonic results of which have been evident in cinemas since the late 1970s and the encoding for which, in the form of Dolby 5.1 on the soundtracks of DVDs, since the turn of the century has been fairly ubiquitous. By way of background, the essay deals in turn with the physical nature of three-dimensional listening and with the history of stereophonic sound as manifest both in the cinema and on LP recordings. More to the point, the essay deals with the aesthetic differences (not just perceptual but also affective) between listening to three-dimensional sounds in real life situations and listening to re-creations of those sounds, via a Dolby system or otherwise, in the privacy and comfort of one’s home. Playing on the homophonic adjectives in its title, the essay reflects on why sometimes we give more rapt attention to artificial versions of “surround sound” than to the genuine stereophonic sound in which we are literally wrapped almost on a daily basis.

Active filters and phase-locked loop (PLL) and its applications are discussed in this chapter. To get acquainted with the design of active filters and the applicability in instrumentation, low pass, high pass and band pass filters are explained. The frequency selection in audio or music systems is utmost important. The selection of lower and higher cut-off frequencies decides the quality of the filters. And in public address or audio or stereophonic system bass and treble decides the quality of the sound and this comes from the proper selection of low, mid and high frequencies.

Traditional stereophonic acoustic echo cancellation algorithms need to estimate acoustic echo paths from stereo loudspeakers to a microphone, which often suffers from the nonuniqueness problem caused by a high correlation between the two far-end signals of these stereo loudspeakers. Many decorrelation methods have already been proposed to mitigate this problem. However, these methods may reduce the audio quality and/or stereophonic spatial perception. This paper proposes to use a convolutional recurrent network (CRN) to suppress the stereophonic echo components by estimating a nonlinear gain, which is then multiplied by the complex spectrum of the microphone signal to obtain the estimated near-end speech without a decorrelation procedure. The CRN includes an encoder-decoder module and two-layer gated recurrent network module, which can take advantage of the feature extraction capability of the convolutional neural networks and temporal modeling capability of recurrent neural networks simultaneously. The magnitude spectra of the two far-end signals are used as input features directly without any decorrelation preprocessing and, thus, both the audio quality and stereophonic spatial perception can be maintained. The experimental results in both the simulated and real acoustic environments show that the proposed algorithm outperforms traditional algorithms such as the normalized least-mean square and Wiener algorithms, especially in situations of low signal-to-echo ratio and high reverberation time RT60.

In recent years, the reversible data hiding (RDH) based on dual stego cover is developing rapidly because of its high capacity and low distortion. For image case, however, two consecutive images of the same image will draw the attention of adversaries during transmission. In this article, we propose a high-capacity RDH scheme using dual-channel audio, by exploiting the natural dual-channel property of the stereo audio. Specifically, we first convert secret message into novenary digits, which could increase the embedding capacity. Then, the magic matrix is used to embed the secret digit into a single-channel audio to generate two single-channel stego-audio. Finally, the two single-channel stego-audio is combined as a convention dual-channel audio. Extensive experiments have demonstrated that our proposed method could significantly boost the stego quality (the SNR is improved by 16% on average), when comparing with the state-of-the-art methods.

The text (continuation of the article Vestnik VGIK, No. 3 (41), 2019). deals with the ideas of S.M. Eisenstein, presented by him in the theoretical work On Stereoscopic Films, in the context of the subsequent development of cinema phonography and modern scientific and theoretical discussions on the problems of correlation of technological and aesthetic aspects of cinema art. Eisensteins article focuses on and analyzes new and controversial technical achievements of cinema, but the authors thoughts reach a high level of understanding the history and prospects of the development of cinema and art in general as an expression of the organic human need for creative and artistic activity. Turning to the history of the theater as the forerunner of cinema, Eisenstein emphasizes the moment of separation of the actor and stage action from the audience. The director sees the technical possibilities of stereo cinema as a way of returning the viewer to the space of direct co-action, complete immersion in the artistic space, integration with the artistic image. Modern multichannel sound technologies have come close to fulfilling Eisensteins dream of drawing the viewer into the screen space and merging it with the artistic creation. But when and to what extent is the use of stereo effects justified, how do the technological and artistic aspects of film production correlate These are issues that are currently the subject of theoretical discourse in the framework of not only film studies, but also interdisciplinary knowledge, affect the development of audiovisual concepts in the practice of filmmaking.

The recent revolution of the Internet as a collaborative medium has opened the door for people who want to share their work. Nevertheless, this may cause serious problems for privacy and copyright protection. Steganography is a powerful tool for protecting important data during transmission. It’s used to hide any secret information like text, image or audio behind a cover file. In this study, a new robust audio steganography technique based on optimum two dimensional Complete Complementary Codes (CCC) has been adopted to encode colour images data and obtain two differently encoded versions of it. These two versions are hidden in DWT coefficients of the two channels of stereo audio signal and embedding locations are determined via 2-D chaotic map random sequence. Complete Complementary Codes (CCC) are sets of spread spectrum sequence family that have ideal auto and cross-correlation properties so, they found many applications in several science areas with the broadest application possibilities in telecommunications. Various attacks are applied to the host audio signals and simulation results show high robustness and capacity with good quality of the extracted image.

“Music Memoirs from Africa” seeks to highlight how music can be used as a tool by contemporary authors to dynamically communicate modern and refreshed experiences of diaspora, sexuality, pan-Africa...

This paper addresses the stereophonic acoustic echo cancelation (SAEC) problem in teleconferencing system by adaptive filtering algorithm. We propose a new stereophonic version of the fast normalized least mean square (FNLMS) algorithm. The proposed algorithm is an extended version of the FNLMS algorithm to the stereophonic case with important improvement of the prediction part that becomes stable in the stereophonic case. The basic idea behind the proposed stereophonic fast normalized least mean square (SFNLMS) algorithm is the normalization of the predictor parameter by the inputs variance for each channel. Simulation results of a comparison between the proposed SFNLMS algorithm and the classical SNLMS version in term of convergence speed and stability are presented.

Acoustic scene classification (ASC) is an important branch of machine hearing. Since ASC systems are intended to be deployed on mobile devices, how to ensure the performance under low-complexity implementation has become an attracting research problem. The state-of-the-art methods include compressing parameter precisions, reducing quantization bits, introducing sparsity constraints and so on. These methods mainly focus on the model level optimization, while explorations are rarely originated from the data level. This paper introduces a train of thoughts from data level, inspired by a stereo audio processing algorithm, namely the primary ambient extraction (PAE), which generates additional samples through audio up-mixing. The experiment results demonstrate that the proposed method exhibits better performance than a group of ASC baseline systems without data level optimization, not to mention that the proposed method is compatible with the existing model level optimization.

The electronic components such as loudspeaker and amplifier in hand-free devices will add the nonlinear distortion (ND) into the echo path of the stereophonic acoustic systems. The stereophonic acoustic echo cancellation (SAEC) algorithms developed to minimize the effect of acoustic echo in these systems assuming linear echo path will suffer from degraded performance in presence of ND. However, no solutions were proposed to address issue of ND in SAEC. This paper introduces a sub-filter based NAEC framework for faster convergence and enhanced steady-state performance. The proposed framework employs the functional link approach in modeling the ND and focuses on improving the rate of convergence of NAEC for better echo cancellation performance. Simulations indicate that the proposed NAEC outperforms the variants of functional link based NAEC in terms of echo return loss enhancement for speech and colored noise inputs.

Stereophonic audio, especially binaural audio, plays an essential role in immersive viewing environments. Recent research has explored generating visually guided stereophonic audios supervised by multi-channel audio collections. However, due to the requirement of professional recording devices, existing datasets are limited in scale and variety, which impedes the generalization of supervised methods in real-world scenarios. In this work, we propose PseudoBinaural, an effective pipeline that is free of binaural recordings. The key insight is to carefully build pseudo visual-stereo pairs with mono data for training. Specifically, we leverage spherical harmonic decomposition and head-related impulse response (HRIR) to identify the relationship between spatial locations and received binaural audios. Then in the visual modality, corresponding visual cues of the mono data are manually placed at sound source positions to form the pairs. Compared to fully-supervised paradigms, our binaural-recording-free pipeline shows great stability in cross-dataset evaluation and achieves comparable performance under subjective preference. Moreover, combined with binaural recordings, our method is able to further boost the performance of binaural audio generation under supervised settings., Accepted by CVPR 2021. Code, models, and demo video are available on our webpage: \

Conveying directional feedback is important for individuals who are blind or have limited visual acuity. However, most studies have focused on supporting two-dimensional guidance. In this work, we investigated the effects of different nonvisual feedback conditions for providing directional guidance in a three-dimensional space. We conducted a user study with six people who are blind or have low vision to investigate the effects of stereo sound (on vs. off) and feedback modalities (beeping vs. vibration vs. beeping+vibration). Participants were asked to point a series of virtual targets randomly appeared around them in 3D with a laser pointer as quickly as possible. Findings suggest that the presence of beeping sound have better performance in terms of task completion time and travel distance compared to when vibration feedback was provided without beeping sound, which was the least preferred condition. In addition, we found that the presence of stereo sound has no significant effect on the performance although it is preferred by most participants. This work can contribute to 3D navigation for people who are blind or have limited visual acuity.

Audio source separation is a cornerstone problem for researchers engaged in Digital Signal Processing and Artificial Intelligence. Music unmixing is the task of decomposing music into its constitutive components, like yielding separated stems for the vocals, bass, drums, accompaniment, jazz, and others from a mastered song track. Due to recent progress in the field of Deep Learning, researchers have been able to devise Neural Networks that can perform this task with considerable precision. However, these models lack performance when dealing with generic musical audio despite having decent utility over specific music genres. The proposed system aims to develop a universal platform-independent software for accurate domain-specific implementation of music source separation for acute subsets of stereo audio using the Bidirectional Long Short Term Memory (BLSTM) architecture of Recurrent Neural Networks. The Deep Neural Network helps demix audio mixtures into the jazz solo and its accompaniment. In cohesion, these two models extract five independent audio stems from the original audio with reasonable accuracy. Further, the extracted accompaniment stem is processed using ConvNet Model to estimate the instrumental components. In synchronization, these three models can break down audio to its fundamental elements.

This chapter covers the fairly dramatic transition period from before to after the war. It is interesting to note that stereophonic sound was invented 8 years before the war but was not used until the 1950s. Early forms of recording sound are discussed as well as the introduction of the tape recorder.

This chapter simply covers how to do things properly by recording true stereo sound, which was invented by Alan Blumlein in 1931, although not properly used before the 1950s. This is mostly because, until the advent of the tape recorder, there was no easy way to record stereophonic sound. Blumlein had managed it in the laboratory, but then, he was a genius. He did define clearly the methodologies for recording stereo, even though the technology was not there at the time. In the 1950s and into the very early 1960s, audio engineers in the classical music field followed these principles and made some wonderful, true stereo recordings, which we may still hear today. But then most of them were overtaken by multi-track technology, and they forgot the art. This chapter shows you how to do it again and produce true stereo recordings.

This chapter introduces the name Alan Dower Blumlein and the tremendous flow of ideas that came from this one man within a decade. It discusses the invention of stereophonic sound, the moving coil microphone and the sound for film. The sudden flourish of what became known simply as ‘High Fidelity’ is introduced and the resultant global market that was developed. Blumlein’s work on RADAR is also briefly discussed as many may not know of this, or his subsequent death while testing new variations of RADAR in an aircraft crash.

The pressure upon recording studios to offer 24-track recording was very real, and many had to invest significantly in not only the 24-track machines but compatible audio mixers and outboard electronics. The installation of such equipment into an existing studio required an enormous effort and associated cost. This was driven mostly by the popular music fraternity, but would come to affect the classical music field as well, with consequent changes in the way classical music was recorded and the sound that record buyers heard, which did not necessarily correspond with the sound of the orchestra as heard in the concert hall. The importance of true stereo sound is also discussed.

It is a fact that many classical music recordings that were made in the late 1950s and early 1960s actually sound much more realistic and are of better quality than many of those made today. This is principally because they were recorded in true stereo with good quality microphones and relatively little in the signal chain. Some audio engineers continue to work this way, but not many. The art and science have largely been lost. In the popular music field, they never really adopted stereo anyway, except from some recordings made in the early 1950s, and so things have not changed much, except of course from the impact of digital technology. These issues are discussed.

The effect of multi-track recording and digital sound caused engineers within the classical music field to break with tradition and start to put microphones in front of almost every instrument, in the knowledge that they could mix the audio in any way they wished after the event. They would argue that this produced the right sound of each instrument, but it did not reproduce the sound of the orchestra as the conductor and audience heard it, or how the composer intended it to be heard. This has introduced a dichotomy within the recording industry with some freelance engineers and specialist labels recording in true stereo, thus preserving the sound of the orchestra in place, and most of the larger labels recording the multi-track way, emulating those within the popular music field. This is discussed in some depth.

While directional guidance is essential for spatial navigation, little has been studied about providing nonvisual cues in 3D space for individuals who are blind or have limited visual acuity. To understand the effects of different nonvisual feedback for 3D directional guidance, we conducted a user study with 12 blind-folded participants. They were asked to search for a virtual target in a 3D space with a laser pointer as quickly as possible under 6 different feedback designs varying the feedback mode (beeping vs. haptic vs. beeping+haptic) and the presence of a stereo sound. Our findings show that beeping sound feedback with and without haptic feedback outperforms the mode where only haptic feedback is provided. We also found that stereo sound feedback generated from a target significantly improves both the task completion time and travel distance. Our work can help people who are blind or have limited visual acuity to understand the directional guidance in a 3D space.

Frequency modulation (FM) is used worldwide for high-fidelity broadcast radio communication. This paper presents a MATLAB implementation of a stereophonic FM transmitter and receiver. Simulations are performed to measure the performance of the stereophonic FM receiver in the presence of additive white Gaussian noise (AWGN). A stereophonic FM signal with pre-detection SNR values from 0 to 60 dB, in increments of 10 dB, is demodulated using a discriminator. Post-detection SNR values are determined, and performance characteristics are examined for above and below threshold modes of FM operation. The simulations show that the receiver reliably demodulates the message signal for pre-detection SNR values greater than 30 dB. Degradation of the signal occurs at and below 20 dB with a complete loss of signal at 0 dB (below threshold mode). A 15 dB post-detection SNR gain occurs at 60 dB pre-detection SNR.

Multi-channel audio systems are widely used in modern sound devices for movie theaters and home theaters. However, most audio files are encoded into stereo tracks. When using a multichannel system to play a stereophonic audio, the system may not produce the best possible listening perception. This study aims to upmix the stereophonic sound to 5.1-channel sound, thereby enhancing surround sound experience. The proposed system extracts the primary sound for the center audio channel and the ambient sound for the side audio channels. The frontal audio channels are reinforced by a special equalization technique. Users can use a smartphone to remotely control the proposed playback. The subjective listening results show that the spatial audio quality outperforms the commercial built-in upmixing especially in spatial hearing. We also hypothesize that there is a trade-off between directional enhancement and spectral balance in the pilot experiment.

Music retrieval methods are in the focus of recent interest due to the increasing size of music databases as e.g. in the Internet. Among different query methods content-based media retrieval analyzing intrinsic characteristics of the source seems to form the most intuitive access. The key-melody in a song can be regarded as the major characteristic in music and leads to a query by humming or singing. In this paper we turn our attention to both, the features and the algorithm of matching in audio music retrieval. Nowadays approaches propagate the use of dynamic time warping for the matching process. As reference mostly midi-data or humming itself is used. However, first attempts matching humming to polyphonic audio exist. In this contribution we introduce hidden Markov models as an alternative for humming queries matching humming itself, mobile phone ring tones and polyphonic audio. The second object of our research is the introduction of a new way of melody enhancement prior to a latter feature extraction by use of stereophonic information. Further an adaptation throughout the extraction process of the frame length to the tempo of a musical piece helps improving similarity matching performance. The paper addresses the design of a working recognition engine and results achieved with respect to the alluded methods. A test database consisting of polyphonic audio clips, ring tones, and sung user data is described in detail.

The recent resurgence of vinyl music records sales led by a consumer demand is increasing faster than production capability. This has resulted in supply delays across the sector. Thus far, manufacturing investments have been focused on traditional proven methods rather than alternative technologies. This paper demonstrates for the first time the production of a stereo recording via analog methodology using a single pulsed laser beam. Using mid-side theory,to combine a sum (mono) signal with a difference signal, a 532nm Nd:Yag laser beam was used to process high-impact polystyrene discs (HIPS). Stereo recordings were manufactured by varying the laser power to produce a difference signal and deflecting the beam with a mirror mounted galvanometer to produce the sum signal. Upon playback on a conventional turntable, the recordings were analyzed with an oscilloscope and stereo separation was observed. To our knowledge this is the first time a stereo signal has been successfully recorded using a single laser beam. Previous literature has used a single laser beam to achieve mono signals and required significant digital pre-processing of the audio source. This new methodology requires lowerinvestment costs than traditional pressing plants and would make volume-tailored production more affordable.

Acoustic echoes retrieval is a research topic that is gaining importance in many speech and audio signal processing applications such as speech enhancement, source separation, dereverberation and room geometry estimation. This work proposes a novel approach to blindly retrieve the off-grid timing of early acoustic echoes from a stereophonic recording of an unknown sound source such as speech. It builds on the recent framework of continuous dictionaries. In contrast with existing methods, the proposed approach does not rely on parameter tuning nor peak picking techniques by working directly in the parameter space of interest. The accuracy and robustness of the method are assessed on challenging simulated setups with varying noise and reverberation levels and are compared to two state-of-the-art methods.

© 2019 Elsevier Ltd Auditory source width (ASW) is the perceived width of an auditory event of a stimulus, which has complicated relationships with the interaural cross-correlation coefficient, loudness level and frequency. In this paper, the virtual acoustics pointer method is used as the reference signals to investigate the relationship for headphone users. It is found that increasing the loudness level increases the ASW, but with different degrees at different frequencies. The minimum ASW increment appears around 1600 Hz and the maximum occurs around 200 Hz. The average of the ASW increment is approximately 5.4 for 10 phons loudness increment. Decreasing IACC can broaden the ASW, the maximum of the increment of ASW occurs around 400–800 Hz and the average increment is approximately 4.8 for the decrement in IACC of 0.2. A formula is developed by curve fitting to describe the relationship among the three factors and the ASW, which can be used to predict the auditory source width for stereo sound reproduction using headphones and help hearing impaired people perceive more accurate ASW of sound.

A new reversible watermarking algorithm on stereo audio signals is proposed in this article. By utilizing correlations between two channels of audio signal, the authors segment one channel based on another one according to the smoothness. For each segmented sub-host sequence, they estimate its capacity and the corresponding embedding distortion firstly, and then select the optimal combinations of sub-host sequences for embedding. Experimental results indicate that the proposed algorithm can improve SNR (signal to noise ratio) for various kinds of capacity.

We investigate the effects of auditory stimuli that improve the ease of touching a virtual object outside a user’s field of view. Our impression evaluation experiments show that a 45-degree expanded stereophonic sound contributes to the natural interaction between a user and a virtual object without using visual information.

The purpose of the work done was to estimate the human binaural capability of locating stereo sound sources of information signals, especially speech, choir singing, and symphonic music as compared to pulse signals. The measurements were conducted in an average-sized hall where sound was emitted by a stereo system consisting of two loudspeakers with the base width of 4.5 m. Based on the results of analysis of existing methods of binaural sound perception modelling, the estimation was performed based on correlational processing of signals recorded using a head dummy placed in various points of the room. For the recording points selected, the maximum time delays between signals arriving at a listener’s left and right ears were calculated, and the interaural cross-correlation functions were obtained. The general functions, especially the peak shift, correlation interval, and peak sharpness (including the conditions when it splits into two individual ones), were analysed. The correlation factor values and levels were calculated. As a result, the conclusions regarding the capability of locating an imaginary sound source were made based on the cross-correlation function factors values, which simplified the application of this method in practice. Based on the results of experiments conducted and on the subjective sensations of perception, the authors have come to the conclusion that a stereophony area has sizes much wider than those assumed earlier from the signal time delay of 1 ms at a receiving stereo pair (representing a shift of an imaginary source towards the ear perceiving the signal earlier). At the same time, it was found that signals with a speech component are much harder to locate than pulse signals. While music signals, especially symphonic music, are close to pulse signals in terms of human locating capabilities. The found patterns have allowed us to introduce adjustments to the stereophony area calculations. Based on the research results, we suggest defining the stereophony zone border by the correlation factor value of 0.5. Given that, the interaural cross-correlation function properties and subjective perception provide for acceptable speech legibility and music transparency. The key conclusion is that this area is quite narrow for speech signals and is actually limited to the time delay of signals arriving at left and right ears – around 1 ms. Music signals have a wider stereophony area defined by the time delay between perception binaural pair components of around 10 ms. Therefore, sizes of a stereophonic sound area ought to be defined with regard to an information signal’s type. Ref. 14, fig. 3, tabl. 3.

Event detection of indoor objects, including doors, has a wide variety of applications, including intruder detection, HVAC control, and surveillance of independently living elderly people. Hence, this has been the focus of multiple research projects in the UbiComp research community. Herein, we propose a method to accurately detect door events in an indoor environment, without the installation and maintenance costs of using distributed ubiquitous sensors. In particular, we recognize the events of multiple doors existing in the environment via active sound probing using a disused smartphone installed in the environment. We perform event recognition by fusing the analysis of the Doppler shift caused by the moving doors with the acoustic characteristics describing the open/close states of the doors acquired via impulse response. To accurately distinguish between the events of different doors via sound probing, our method employs the time-series analysis of the Doppler shift as well as the active sound probing using directional high-frequency sine waves and stereo sound recording. In addition, by incorporating prior knowledge about the state transitions of a door object into a recognition model, we attempt to improve the accuracy of event recognition. Moreover, our method is capable of recognizing walking activities of a person related to door events in the environment, which are necessary information for applications such as HVAC control that require information about both door events and human presence.

Auditory spatial perception relies on more than one spatial cue. This study investigated the effects of cue congruence on auditory localization and the extent of visual bias between two binaural cues-interaural time differences (ITDs) and interaural level differences (ILDs). Interactions between these binaural cues were manipulated by stereophonic techniques. The results show that incoherent binaural information increased auditory response noise and amplified visual bias. The analysis further suggests that although ILD is not the dominant cue for low-frequency localization, it may strengthen the position estimate by combining with the dominant ITD information to minimize estimation noise.

In this paper, a unified framework for using speech interfaces with intelligent televisions (TVs) is proposed. The proposed framework is primarily based on suppressing the stereophonic acoustic echo of the TV audio sounds and adaptive tracking of the time varying acoustic transfer function (ATF) of the echo sounds. In order to effectively suppress the acoustic echoes from the TV speakers, an optimal filter with a multichannel subspace approach is imposed on noisy inputs and the second order statistics of the stereophonic echoes are adaptively estimated by tracking the time varying ATF using the normalized least mean square method. Through assessing the proposed method with a database collected in real TV watching environments, noticeable improvements were achieved in the signal to noise ratio gain, Itakura–Saito distance, and speech recognition rates.

This paper proposes a blind digital audio water- marking algorithm that utilizes the quantization index modulation (QIM) and the singular value decomposition (SVD) of stereo audio signals. Conventional SVD-based blind audio watermarking algorithms lack physical interpretation since the matrix construction method for the input matrix for SVD is heuristically defined. However, in the proposed approach, because the SVD is directly applied to the stereo input signals, the resulting decomposed elements convey a conceptually meaningful inter- pretation of the original audio signal. As the proposed approach effectively utilizes the ratio of singular values, the embedded watermark is highly imperceptible and robust against volumetric scaling attacks; most QIM-based watermarking schemes are weak to these types of attacks. Experimental results under well-known practical attacks, such as compressions, resampling, and various types of signal processing, confirm that the proposed algorithm performs well compared to conventional audio watermarking algorithms.

Stereophonic reproduction is an art form often at odds with analytical testing. The author presents a breakdown of the aspects of stereo reproduction, including musical acoustics and the electroacoustics, to room acoustic assessment and the psychoacoustics used in the assessment and enjoyment of this common art form. This goal is to provide a holistic overview and encourage discussion and improvements in the processes used in the production of the art, as well as the analysis and dialog of the consumers of the art.

We explore the properties of sound and human sound recognition as a means to enhance and accelerate visual-only data analysis methods. The aim of this work is to enable and improve the analysis of large data sets, data requiring rapid analysis, multi-dimensional data, and signal detection in data with low signal-to-noise ratio. We present a prototype tool, StarSound, to sonify data such as astronomical transient light curves, spectra, and power spectra. Stereophonic sound is used to ‘visualise’ and localise the data under examination, and 3-D sound is discussed in conjunction with virtual reality technology, as a means to enhance analysis efficiency and efficacy, including rapid data assessment and training machine learning software. In addition, we explore the use of higher-order harmonics as a means to examine simultaneously multi-dimensional data sets. Such an approach can allow the data to be interpreted in a holistic manner and facilitates the discovery of previously unseen connections and relationships. Furthermore, we exploit the capability of the human brain for selective or focused hearing that enables the identification of desired signals in noisy data, or amidst similar or more significant signals. Finally, we provide research examples that benefit directly from data sonification. The work presented here aims to help tackle the challenges of the upcoming era of Big Data and help optimise, speed up and expand aspects of data analysis requiring human interaction.

As technology develops in today’s age, there are many radio equipment easily found in the market. Certainly, there is a stereo decoder module inside a FM radio receiver which operates to return radio signal that previously was modulated and broken apart into stereo audio signal (L+R) and (L-R). There are many stereo decoders already presented inside FM radio receivers. Based on this premise, there is a problem that can be studied, which is how to design a decoder stereo in a stereo FM radio receiver that functions as the demodulation in stereo FM radio and operates to release L+R audio signal to a receiver radio. Therefore, it needs a signal demodulatorof encoder stereo that converts L and R audio signal. Based on the testing result, it showed that sideband frequency from the result of DSB-SC demodulator process, which was produced through R and L channels, still had frequency value differences with average percentage of 1.416% out of the expected frequency value. This was due to the frequency oscillator from the generator function.

This article studies the effects of inter-channel time and level differences in stereophonic reproduction on perceived localization uncertainty, which is defined as how difficult it is for a listener to tell where a sound source is located. Towards this end, a computational model of localization uncertainty is proposed first. The model calculates inter-aural time and level difference cues, and compares them to those associated to free-field point-like sources. The comparison is carried out using a particular distance functional that replicates the increased uncertainty observed experimentally with inconsistent inter-aural time and level difference cues. The model is validated by formal listening tests, achieving a Pearson correlation of 0.99. The model is then used to predict localization uncertainty for stereophonic setups and a listener in central and off-central positions. Results show that amplitude methods achieve a slightly lower localization uncertainty for a listener positioned exactly in the center of the sweet spot. As soon as the listener moves away from that position, the situation reverses, with time-amplitude methods achieving a lower localization uncertainty.

Environmental Sound Classification (ESC) is an active research area in the audio domain and has seen a lot of progress in the past years. However, many of the existing approaches achieve high accuracy by relying on domain-specific features and architectures, making it harder to benefit from advances in other fields (e.g., the image domain). Additionally, some of the past successes have been attributed to a discrepancy of how results are evaluated (i.e., on unofficial splits of the UrbanSound8K (US8K) dataset), distorting the overall progression of the field. The contribution of this paper is twofold. First, we present a model that is inherently compatible with mono and stereo sound inputs. Our model is based on simple log-power Short-Time Fourier Transform (STFT) spectrograms and combines them with several well-known approaches from the image domain (i.e., ResNet, Siamese-like networks and attention). We investigate the influence of cross-domain pre-training, architectural changes, and evaluate our model on standard datasets. We find that our model out-performs all previously known approaches in a fair comparison by achieving accuracies of 97.0 % (ESC-10), 91.5 % (ESC-50) and 84.2 % / 85.4 % (US8K mono / stereo). Second, we provide a comprehensive overview of the actual state of the field, by differentiating several previously reported results on the US8K dataset between official or unofficial splits. For better reproducibility, our code (including any re-implementations) is made available., Comment: 8 pages, 4 figures; submitted to ICPR 2020

Stereophonic audio is an indispensable ingredient to enhance human auditory experience. Recent research has explored the usage of visual information as guidance to generate binaural or ambisonic audio from mono ones with stereo supervision. However, this fully supervised paradigm suffers from an inherent drawback: the recording of stereophonic audio usually requires delicate devices that are expensive for wide accessibility. To overcome this challenge, we propose to leverage the vastly available mono data to facilitate the generation of stereophonic audio. Our key observation is that the task of visually indicated audio separation also maps independent audios to their corresponding visual positions, which shares a similar objective with stereophonic audio generation. We integrate both stereo generation and source separation into a unified framework, Sep-Stereo, by considering source separation as a particular type of audio spatialization. Specifically, a novel associative pyramid network architecture is carefully designed for audio-visual feature fusion. Extensive experiments demonstrate that our framework can improve the stereophonic audio generation results while performing accurate sound separation with a shared backbone (Code, models and demo video are available at https://hangz-nju-cuhk.github.io/projects/Sep-Stereo.).

This article presents the design of a tangible tool for teaching basic programming concepts to students with visual impairments and low vision. Meeting the preliminary requirements of this ongoing project (Tangible input, cost-efficient, and minimal maintenance), the paper describes the design of the hardware of this prototype with inexpensive materials, such as Lego blocks, 3.5 mm stereo audio jack connectors, relays, resistors, and an Arduino Mega microcontroller. To identify the Lego blocks that represent the code instructions, the system provides three different methods of identification; color, 3D printed labels, and braille labels. To acquire feedback and validate this prototype, a preliminary study with nine participants with Visual Impairment and Low Vision has been carried out. All participants completed the task successfully, provided feedback about the prototype, and all recommended this prototype to teach basic programming concepts. Currently, the project is focused on the implementation of music as an output method of programs created by users. The objective is to create a programming tool completely independent of visual graphics or information.