Is Anyone Listening? An Examination of New Zealand Musicians in the Digital Age

<p>With major recording artist Thom Yorke predicting the record industry will crumble in “Months” (Hudson, 2010), and sensationalist headlines such as “iPods and Young People Have Utterly Destroyed Music” (Buchanan, 2009) becoming commonplace, this research attempts to determine the current state of New Zealand music in the digital age. Despite the doom and gloom coming from the press in regards to the music industry, musicians haven’t stopped continuing to record, release, and promote their music as the costs of doing so continues to decline with the advent of new technologies. This research looks specifically into the music hosting website Bandcamp and determines what methods New Zealand musicians are currently using on the site in an effort to get their music into the ears and onto the hard drives of fans. Although a large amount of research has been performed on the impacts of piracy on music sales, very little has been conducted on what strategies musicians are implementing to increase their exposure and connect with their fan base in the 21st century, with no specific research having been performed on the unique circumstances faced by artists in New Zealand. This paper first presents a historical overview of the music industry in the last century, as well as a summary of where the industry currently stands in regards to Copyright, distribution methods, and price models in order to provide perspective on the difficulties and variety of choices currently facing musicians. Within this research paper, several hypotheses were tested in order to determine what factors have a significant effect on the amount of exposure that an artist has received for their music. In order to test these hypotheses, the number of audio streams and downloads that an artist has received for their songs posted to the music hosting site Bandcamp was used as a measure to determine the amount of exposure that a specific artist has received. Due to the subjective nature of the quality of music which each musician creates, a survey was sent to over 500 New Zealand musicians whom provided at least one song for download on the website in order to gather as much overall data on the success generated by New Zealand musicians online as possible. A quantitative analysis was then performed to determine what social networking and music hosting sites are most popular with Kiwi artists; whether musicians are still creating physical copies of their works; and what licenses and payment models artists are applying to their songs. This analysis identified two important factors as statistically significant in terms of affecting the number of downloads and audio streams an artist receives on Bandcamp, the length of time that an artist has been present on the site and the payment model that an artist applies to their works. In addition to the quantitative analysis performed on the success that artists were achieving on Bandcamp, a qualitative analysis was performed on the motivations artists had for applying specific pricing models and licenses to their works. The results of this analysis found a nearly unanimous positive response from musicians who had applied traditional Copyright to their work when asked if they would allow their fans to share their music without expressed permission. This research also determined that a majority of musicians currently applying traditional Copyright to their works are unfamiliar, unaware, or uninformed about Creative Commons licenses, with traditional Copyright being applied more out of habit than a desire for their works to be protected under the rights granted under traditional Copyright. A discussion about what these results indicate for artists is also presented as a guide for future and current musicians looking to upload their music to Bandcamp, depending on the goals that the musician is looking to achieve with their music. Finally, this paper concludes with an analysis of what limitations are present in the results of the research, as well as where the need exists for future research.</p>

Is Anyone Listening? An Examination of New Zealand Musicians in the Digital Age

<p>With major recording artist Thom Yorke predicting the record industry will crumble in “Months” (Hudson, 2010), and sensationalist headlines such as “iPods and Young People Have Utterly Destroyed Music” (Buchanan, 2009) becoming commonplace, this research attempts to determine the current state of New Zealand music in the digital age. Despite the doom and gloom coming from the press in regards to the music industry, musicians haven’t stopped continuing to record, release, and promote their music as the costs of doing so continues to decline with the advent of new technologies. This research looks specifically into the music hosting website Bandcamp and determines what methods New Zealand musicians are currently using on the site in an effort to get their music into the ears and onto the hard drives of fans. Although a large amount of research has been performed on the impacts of piracy on music sales, very little has been conducted on what strategies musicians are implementing to increase their exposure and connect with their fan base in the 21st century, with no specific research having been performed on the unique circumstances faced by artists in New Zealand. This paper first presents a historical overview of the music industry in the last century, as well as a summary of where the industry currently stands in regards to Copyright, distribution methods, and price models in order to provide perspective on the difficulties and variety of choices currently facing musicians. Within this research paper, several hypotheses were tested in order to determine what factors have a significant effect on the amount of exposure that an artist has received for their music. In order to test these hypotheses, the number of audio streams and downloads that an artist has received for their songs posted to the music hosting site Bandcamp was used as a measure to determine the amount of exposure that a specific artist has received. Due to the subjective nature of the quality of music which each musician creates, a survey was sent to over 500 New Zealand musicians whom provided at least one song for download on the website in order to gather as much overall data on the success generated by New Zealand musicians online as possible. A quantitative analysis was then performed to determine what social networking and music hosting sites are most popular with Kiwi artists; whether musicians are still creating physical copies of their works; and what licenses and payment models artists are applying to their songs. This analysis identified two important factors as statistically significant in terms of affecting the number of downloads and audio streams an artist receives on Bandcamp, the length of time that an artist has been present on the site and the payment model that an artist applies to their works. In addition to the quantitative analysis performed on the success that artists were achieving on Bandcamp, a qualitative analysis was performed on the motivations artists had for applying specific pricing models and licenses to their works. The results of this analysis found a nearly unanimous positive response from musicians who had applied traditional Copyright to their work when asked if they would allow their fans to share their music without expressed permission. This research also determined that a majority of musicians currently applying traditional Copyright to their works are unfamiliar, unaware, or uninformed about Creative Commons licenses, with traditional Copyright being applied more out of habit than a desire for their works to be protected under the rights granted under traditional Copyright. A discussion about what these results indicate for artists is also presented as a guide for future and current musicians looking to upload their music to Bandcamp, depending on the goals that the musician is looking to achieve with their music. Finally, this paper concludes with an analysis of what limitations are present in the results of the research, as well as where the need exists for future research.</p>

Digitally-Signed Video/Audio Streams as Prevention of AI-Based Attacks

The main purpose of the research is to provide the solution that allows digitally signing video and audio records, which is extremely important not only in the frame of copyright protection but also when it comes to understanding whether a video is fake. In the world where neural networks are used more and more often, it becomes easy to make a fake video using the face of some famous politician, maybe even the president of the US, and broadcast such a video, causing different kinds of negative events in the situation of the political crises.

Audio Information Camouflage Detection for Social Networks

Sending camouflaged audio information for fraud in social networks has become a new means of social networks attack. The hidden acoustic events in the audio scene play an important role in the detection of camouflaged audio information. Therefore, the application of machine learning methods to represent hidden information in audio streams has become a hot issue in the field of network security detection. This study proposes a heuristic mask for empirical mode decomposition (HM-EMD) method for extracting hidden features from audio streams. The method consists of two parts: First, it constructs heuristic mask signals related to the signal’s structure to solve the modal mixing problem in intrinsic mode function (IMF) and obtains a pure IMF related to the signal’s structure. Second, a series of hidden features in environment-oriented audio streams is constructed on the basis of the IMF. A machine learning method and hidden information features are subsequently used for audio stream scene classification. Experimental results show that the hidden information features of audio streams based on HM-EMD are better than the classical mel cepstrum coefficients (MFCC) under different classifiers. Moreover, the classification accuracy achieved with HM-EMD increases by 17.4 percentage points under the three-layer perceptron and by 1.3% under the depth model of TridentResNet. The hidden information features extracted by HM-EMD from audio streams revealed that the proposed method could effectively detect camouflaged audio information in social networks, which provides a new research idea for improving the security of social networks.

Feature Extraction Method for Hidden Information in Audio Streams Based on HM-EMD

Using fake audio to spoof the audio devices in the Internet of Things has become an important problem in modern network security. Aiming at the problem of lack of robust features in fake audio detection, an audio streams’ hidden feature extraction method based on a heuristic mask for empirical mode decomposition (HM-EMD) is proposed in this paper. First, using HM-EMD, each signal is decomposed into several monotonic intrinsic mode functions (IMFs). Then, on the basis of IMFs, basic features and hidden information features HCFs of audio streams are constructed, respectively. Finally, a machine learning method is used to classify audio streams based on these features. The experimental results show that hidden information features of audio streams based on HM-EMD can effectively supplement the nonlinear and nonstationary information that traditional features such as mel cepstrum features cannot express and can better realize the representation of hidden acoustic events, which provide a new research idea for fake audio detection.

A positive role of task-deactivated default mode network in narrative speech comprehension

Psychological theories have implicated an active role of the default mode network (DMN) in natural speech comprehension. However, as listeners need to keep tracking the external audio streams, the DMN is regularly de-activated and anticorrelated with externally-oriented networks. Such a pattern has been interpreted as the suppression of the DMN to support externally-oriented cognitive processes. The current study aims to resolve this seeming contradiction. Brain activities from a speaker telling autobiographical stories and a group of participants (N = 62) listening to the recordings were collected with fMRI. By analyzing the listeners’ brains alone, we found the DMN was deactivated during speech listening relative to a fixation period and anticorrelated with the task-positive perisylvian language network (pLN). Dynamic Causal Modeling showed the pLN had inhibitory influence on the DMN, whereas the DMN had excitatory influence on the pLN. Further between-brain analyses revealed the activities of DMN in the listener’s brain were tightly coupled with the activities of the homologous network in the speaker’s brain. Significant interbrain couplings were also observed in the pLN, but were weaker and faded quicker. Moreover, listeners showing stronger coupling responses to the speaker in the DMN understood the speech better, and tended to exhibit more positive DMN → pLN effective connections. We conclude that the DMN may occupy an internal system that works cooperatively with the externally-oriented pLN to support narrative speech comprehension.

Deep multiple instance learning for foreground speech localization in ambient audio from wearable devices

Over the recent years, machine learning techniques have been employed to produce state-of-the-art results in several audio related tasks. The success of these approaches has been largely due to access to large amounts of open-source datasets and enhancement of computational resources. However, a shortcoming of these methods is that they often fail to generalize well to tasks from real life scenarios, due to domain mismatch. One such task is foreground speech detection from wearable audio devices. Several interfering factors such as dynamically varying environmental conditions, including background speakers, TV, or radio audio, render foreground speech detection to be a challenging task. Moreover, obtaining precise moment-to-moment annotations of audio streams for analysis and model training is also time-consuming and costly. In this work, we use multiple instance learning (MIL) to facilitate development of such models using annotations available at a lower time-resolution (coarsely labeled). We show how MIL can be applied to localize foreground speech in coarsely labeled audio and show both bag-level and instance-level results. We also study different pooling methods and how they can be adapted to densely distributed events as observed in our application. Finally, we show improvements using speech activity detection embeddings as features for foreground detection.

Laser-sound: optoacoustic transduction from digital audio streams

This work presents a novel laser-based optoacoustic transducer capable of reproducing controlled and continuous sound of arbitrary complexity in the air or on solid targets. Light-to-sound transduction is achieved via laser-induced breakdown, leading to the formation of plasma acoustic sources in any desired spatial location. The acoustic signal is encoded into pulse streams via a discrete-time audio modulation and is reproduced by fast consecutive excitation of the target medium with appropriately modulated laser pulses. This results in the signal being directly reconstructed at the desired location of the target medium without the need for a receiver or demodulation device. In this work, the principles and evaluation results of such a novel laser-sound prototype system are presented. The performance of the prototype is evaluated by systematic experimental measurements of audio test signals, from which the basic acoustical response is derived. Moreover, a generic computational model is presented that allows for the simulation of laser-sound reproduction of 1-bit or multibit audio streams. The model evaluations are validated by comparison with the acoustic measurements, whereby a good agreement is found. Finally, the computational model is used to simulate an ideal optoacoustic transducer based on the specifications of state-of-the-art commercially available lasers.