The need for music information retrieval with user-centered and multimodal strategies

Abstract

Music is a widely enjoyed content type, existing in many multifaceted representations. With the digital information age, a lot of digitized music information has theoretically become available at the user's fingertips. However, the abundance of information is too large-scaled and too diverse to annotate, oversee and present in a consistent and human manner, motivating the development of automated Music Information Retrieval (Music-IR) techniques.In this paper, we encourage to consider music content beyond a monomodal audio signal and argue that Music-IR approaches with multimodal and user-centered strategies are necessary to serve reallife usage patterns and maintain and improve accessibility of digital music data. After discussing relevant existing work in these directions, we show that the field of Music-IR faces similar challenges as neighboring fields, and thus suggest opportunities for joint collaboration and mutual inspiration.

The neglected user in music information retrieval research

“…The authors give some examples of such context categories and particular aspects: social and cultural context (political and national context), everyday situations (work, leisure, consumer, entertainment), presence/absence of others (live, audience, recorder). It is exactly this multifaceted and individual way of music perception that has largely been neglected so far when elaborating and evaluating music retrieval approaches, but should be given more attention, in particular considering the trend towards personalized and context-aware systems (Liem et al 2011;Schedl and Knees 2011).…”

Genre-based Analysis of Social Media Data on Music Listening Behavior

“…Mining and analyzing musical preference and other personal characteristics is becoming more and more important to elaborate user-centric music retrieval systems [5]. Musical information gathered from social media, in particular from microblogs and online music platforms, can be used to build or enrich music recommender systems, as demonstrated in [11] and [2], among others.…”

Hybrid retrieval approaches to geospatial music recommendation

“…It is en-titled MusicMicro and is freely available 2 , fostering reproducibility of social media-related MIR research. This data set contains about 600,000 listening events posted on Twitter 3 . Each event is represented by a tuple <twitter-id, userid, month, weekday, longitude, latitude, country-id, city-id, artist-id, track-id>, which allows for spatiotemporal identification of listening behavior.…”

“…Although quite a few MIR researchers suggest such a combination [2,1,3,12,5], a systematic evaluation of combining state-of-the-art audio and web similarity estimators is still missing, hence provided here.…”

“…The field of Music Information Retrieval (MIR) is seeing a paradigm shift, away from system-centric perspectives towards user-centric approaches [3]. In this vein, incorporating user models and addressing user-specific demands in music retrieval and music recommendation systems is becoming more and more important.…”