Online Music Streaming Services Research Articles

EVALUATING DEEP LEARNING MODELS FOR MUSIC EMOTION RECOGNITION

Music listening helps people not only for entertainment, but also to reduce emotional stress in their daily lives. People nowadays tend to use online music streaming services such as Spotify, Amazon Music, Google Play Music, etc. rather than storing the songs on their devices. The songs in these streaming services are categorized into different emotional labels such as happy, sad, romantic, devotional, etc. In the music streaming applications, the songs are manually tagged with their emotional categories for music recommendation. Considering the growth of music on different social media platforms and the internet, the need for automatic tagging will increase in coming time. The work presented deals with training the deep learning model for automatic emotional tagging. It covers implementation of two different deep learning architectures for classifying the audio files using the Mel-spectrogram of music audio. The first architecture proposed is Convolutional Recurrent Model (CRNN) and the second architecture is a Parallel Convolutional Recurrent Model (Parallel CNN). Both the architectures exploit the combined features of Convolutional and Recurrent layers. This combination is used to extract features from time and frequency domains. The results with accuracies in the range of 51 to 54 % are promising for both models for a small dataset of 138 songs, considering the large datasets required for training deep learning models.

DEMYSTIFYING MUSICAL PREFERENCES AT TURKISH MUSIC MARKET THROUGH AUDIO FEATURES OF SPOTIFY CHARTS

Online music streaming services are one of the important actors in music consumption for today’s consumers. In addition to widespread use of mobile devices, many changes in the patterns of music consumption are witnessed such as the purchase of single tracks instead of albums, listening to music on different platforms, and personalized music consumption options. This study aims to examine the concept of music consumption in Turkey through audio characteristics of popular songs. Top 200 popular song-lists for 6 months period are chosen as sample and audio characteristics provided by Spotify API service regarding 676 unique songs are analyzed. Following descriptive statistics of Turkey Music Market, clustering methodology is employed and three different clusters for songs are concluded. Finally, decision tree methodology is employed to classify the dataset with popularity scores and audio characteristics together, while loudness and energy characteristics are found as significant classifiers.

Feature-combination hybrid recommender systems for automated music playlist continuation

Music recommender systems have become a key technology to support the interaction of users with the increasingly larger music catalogs of on-line music streaming services, on-line music shops, and personal devices. An important task in music recommender systems is the automated continuation of music playlists, that enables the recommendation of music streams adapting to given (possibly short) listening sessions. Previous works have shown that applying collaborative filtering to collections of curated music playlists reveals underlying playlist-song co-occurrence patterns that are useful to predict playlist continuations. However, most music collections exhibit a pronounced long-tailed distribution. The majority of songs occur only in few playlists and, as a consequence, they are poorly represented by collaborative filtering. We introduce two feature-combination hybrid recommender systems that extend collaborative filtering by integrating the collaborative information encoded in curated music playlists with any type of song feature vector representation. We conduct off-line experiments to assess the performance of the proposed systems to recover withheld playlist continuations, and we compare them to competitive pure and hybrid collaborative filtering baselines. The results of the experiments indicate that the introduced feature-combination hybrid recommender systems can more accurately predict fitting playlist continuations as a result of their improved representation of songs occurring in few playlists.

The Role of Musical Attributes in Music-Induced Analgesia: A Preliminary Brief Report.

Music-induced analgesia (MIA) is the ability of music to influence pain perception. Although this phenomenon has been extensively studied in recent years, only a few studies have addressed what musical characteristics are optimal for MIA. Here, we present a novel approach to this topic, using a recently proposed model of music attribute preferences. The model addresses three musical dimensions: arousal, valence, and depth. Thirty participants (15 women and 15 men, Mage = 37.1 years, standard deviation = 15.7) were subjected to experimental pain stimulation (cold-pressor task) while listening to brief music excerpts with characteristics of the three attribute dimensions. Each excerpt was selected to score high on one of the three attributes while being average on the other two, to create three distinct music conditions. There was also a control condition, where participants listened to white noise. Results showed that average pain ratings were significantly lower in the arousal (p = 0.002) and depth (p = 0.01) conditions compared to the control condition. Furthermore, participants showed increased pain tolerance in musical conditions compared to the control condition (p = 0.04). This preliminary report introduces a novel approach to studying MIA in the context of music attribute preferences. With the advent of online music streaming services, this research opens new possibilities for music-based pain interventions.

Nothing personal: algorithmic individuation on music streaming platforms.

Raymond Williams once wrote, ‘… there are in fact no masses, but only ways of seeing people as masses’. In an age of personalized media, the word ‘masses’ seems like an anachronism. Nevertheless, if Williams were to study contemporary online platforms, he would no doubt conclude that there are in fact no individuals, but only ways of seeing people as individuals. This article explores this idea by taking a closer look at online music streaming services. It first conducts a comparison of how two leading streaming platforms conceive of the individual music listener. Then, drawing from Gilbert Simondon’s theory of individuation, it demonstrates how ways of seeing the individual work to enact the individual on these platforms. In particular, ways of seeing are heavily influenced by the consumer categories that are defined and demanded by advertisers. This article concludes with an examination of how commercial imperatives shape ‘ways of seeing’ and ‘algorithmic individuation’ on music streaming platforms.

A bipartite fitness model for online music streaming services

This paper proposes an evolution model and an analysis of the behavior of music consumers on online music streaming services. While previous studies have observed power-law degree distributions of usage in online music streaming services, the underlying behavior of users has not been well understood. Users and songs can be described using a bipartite network where an edge exists between a user node and a song node when the user has listened that song. The growth mechanism of bipartite networks has been used to understand the evolution of online bipartite networks Zhang et al. (2013). Existing bipartite models are based on a preferential attachment mechanism László Barabási and Albert (1999) in which the probability that a user listens to a song is proportional to its current popularity. This mechanism does not allow for two types of real world phenomena. First, a newly released song with high quality sometimes quickly gains popularity. Second, the popularity of songs normally decreases as time goes by. Therefore, this paper proposes a new model that is more suitable for online music services by adding fitness and aging functions to the song nodes of the bipartite network proposed by Zhang et al. (2013). Theoretical analyses are performed for the degree distribution of songs. Empirical data from an online streaming service, Last.fm, are used to confirm the degree distribution of the object nodes. Simulation results show improvements from a previous model. Finally, to illustrate the application of the proposed model, a simplified royalty cost model for online music services is used to demonstrate how the changes in the proposed parameters can affect the costs for online music streaming providers. Managerial implications are also discussed.

Critical success factors of online music streaming services - a case study of applying the fuzzy cognitive maps method

The business model of online music streaming allows customers to listen to music on their preferred devices without owning any digital music files. The aim of this paper is to identify which are the critical success factors (CSFs) of online music streaming services, what are the relationships between them and at what level the CSFs influence the business overall success. To achieve this aim, a case-study was applied that utilised the fuzzy cognitive maps (FCMs) method. FCMs offer a soft computing tool capable of capturing human knowledge and behaviour of complex problems in uncertain environments. As a case-study to apply the FCM method we selected Bandcamp, a music streaming service that allows users to stream full music albums for free and purchase music in digital/physical format right from the artist's web page. Bandcamp puts a different spin on the streaming music business model. It is an online music store, a streaming service and an artist promotion platform, all at the same time. The research findings suggest that the core CSFs of online music streaming are free music streaming, the ability to purchase digital and physical format music, the lack of advertisements and the satisfaction of supporting one's favourite artists.

Cannibalizing or Complementing?. The Impact of Online Streaming Services on Music Record Sales

The Internet has changed the media contents industry, and the music industry's business form has transformed from records into online digital products as the online music platform increasingly develops. Before the stabilization of the online music platform, digital music files were considered music piracy and were related to the decline of music record sales. Although research finds that increased Internet penetration led the decline of music record sales, after the stabilization of the online music platform and the introduction of online streaming services in the mid-2000s, a recognizable increase in music record sales appeared. We analyze the impact of adopting online music streaming services on the sales of music records to determine the effect of digital products on physical products by a two-stage least squares (2SLS) regression approach. Using monthly music record sales and streaming data from the official Korean Gaon Music Chart, we find that online streaming services positively impact music record sales. Furthermore, male artists sell more records than female artists as female fans, who prefer male artists, have a higher willingness-to-pay for individual services. Both the number of songs included in one album and the number of artists in one group also positively affect music record sales. Our analysis suggests that online streaming positively impacts music record sales.

Studying emotion induced by music through a crowdsourcing game

One of the major reasons why people find music so enjoyable is its emotional impact. Creating emotion-based playlists is a natural way of organizing music. The usability of online music streaming services could be greatly improved by developing emotion-based access methods, and automatic music emotion recognition (MER) is the most quick and feasible way of achieving it. When resorting to music for emotional regulation purposes, users are interested in the MER method to predict their induced, or felt emotion. The progress of MER in this area is impeded by the absence of publicly accessible ground-truth data on musically induced emotion. Also, there is no consensus on the question which emotional model best fits the demands of the users and can provide an unambiguous linguistic framework to describe musical emotions. In this paper we address these problems by creating a sizeable publicly available dataset of 400 musical excerpts from four genres annotated with induced emotion. We collected the data using an online “game with a purpose” Emotify, which attracted a big and varied sample of participants. We employed a nine item domain-specific emotional model GEMS (Geneva Emotional Music Scale). In this paper we analyze the collected data and report agreement of participants on different categories of GEMS. We also analyze influence of extra-musical factors on induced emotion (gender, mood, music preferences). We suggest that modifications in GEMS model are necessary.

Fighting Free with Free: Freemium vs. Piracy

In this article, we show how new online music streaming services such as Spotify and Deezer can deter piracy. These new services monetize large audiences with low willingness to pay for music. We analyze a simple freemium model in which a firm offers both a free version and a premium version. The firm can restrict the use of the free version. Consumers can choose between the free and the premium version, but can also get a digital copy. More restrictions can increase the number of premium users but diverts other users to piracy. On the contrary, fewer restrictions deter online piracy. We therefore challenge the idea that without strong copyright laws, there is no market for online music. We also argue that there are market solutions to fight free with free that better segment consumers audiences according to their willingness to pay for digital music.

Digital Music Consumption on the Internet: Evidence from Clickstream Data

The goal of this paper is to analyze the behavior of digital music consumers on the Internet. Using clickstream data on a panel of more than 16,000 European consumers, we estimate the effects of illegal downloading and legal streaming on the legal purchases of digital music. Our results suggest that Internet users do not view illegal downloading as a substitute to legal digital music. Although positive and significant, our estimated elasticities are essentially zero: a 10% increase in clicks on illegal downloading websites leads to a 0.2% increase in clicks on legal purchases websites. Online music streaming services are found to have a somewhat larger (but still small) effect on the purchases of digital sound recordings, suggesting complementarities between these two modes of music consumption. According to our results, a 10% increase in clicks on legal streaming websites lead to up to a 0.7% increase in clicks on legal digital purchases websites. We find important cross country difference in these effects.

Vulnerabilities and Countermeasures of Commercial Online Music Streaming Services in Korea

The music industry is rapidly moving from analog to digital, and most of the big portal sites provide commercial online music streaming services. In this paper, we analyze the vulnerabilities of commercial online music streaming services provided by Korea’s major portal sites (Dosirak, Cyworld, and Naver). We show attacks on commercial online music streaming services that lead to an infringement of copyright and propose appropriate countermeasures.

국내 포털사이트 음원서비스 취약점분석 및 대응방안 제안

Nowadays, the music industry is moving from analog to digital. Most of the big portal sites provide commercial online music streaming services. In this paper, we analyzed the security of the Korean commercial online music streaming services which are provided by the Korea's major portal sites(Dosirak, Cyworld, and Naver). We will show attacks on commercial online music streaming services that lead to an infringement of the copyright and we will also propose countermeasures for online commercial music streaming services.