[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-122214-en":3,"doc-seo-122214-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},122214,13056703019404,"Miles","https://ap-avatar.wpscdn.com/davatar_29158cc5080c5b710cf443261637dec0",8,"Research & Report","Reproduction of Black-Box Music Analysis Algorithms through Machine Learning - Diplomarbeit","In the research field of Music Information Retrieval, a set of semantic music features offered by Spotify has gained scientific relevance because these descriptors are used across prior studies. The original computation algorithms are not publicly disclosed, so the features effectively rely on black-box methods. This work analyzes the properties of the investigated music features and studies approaches to reproduce or approximate the underlying black-box algorithms using current Music Information Retrieval methods and machine learning. Features such as Danceability, Acousticness, and Loudness are examined, followed by model-based reproduction and evaluation of fidelity.","Reproduction of Black-Box Music Analysis Algorithms through Machine Learning  \nDIPLOMARBEIT  \nzur Erlangung des akademischen Grades  \nDiplom-Ingenieur  \nim Rahmen des Studiums  \nData Science  \neingereicht von  \nAndreas Schmidt, BSc  \nMatrikelnummer 01526918  \nan der Fakultät für Informatik der Technischen Universität Wien  \nBetreuung: Associate Prof. Dipl.-Ing. Dr.techn. Peter Knees  \nWien, 3 . Juli 2023      \nAndreas Schmidt Peter Knees  \nTechnische Universität Wien A-1040 Wien  Karlsplatz 13  Tel. +43-1-58801-0  [www.tuwien.at](www.tuwien.at)  \nReproduction of Black-Box Music Analysis Algorithms through Machine Learning  \nDIPLOMA THESIS  \nsubmitted in partial fulfillment of the requirements for the degree of  \nDiplom-Ingenieur  \nin  \nData Science  \nby  \nAndreas Schmidt, BSc  \nRegistration Number 01526918  \nto the Faculty of Informatics at the TU Wien  \nAdvisor: Associate Prof. Dipl.-Ing. Dr.techn. Peter Knees  \nVienna, 3rd July, 2023      \nAndreas Schmidt Peter Knees  \nTechnische Universität Wien A-1040 Wien  Karlsplatz 13  Tel. +43-1-58801-0  [www.tuwien.at](www.tuwien.at)  \nErklärung zur Verfassung der Arbeit  \nAndreas Schmidt, BSc  \nHiermit erkläre ich, dass ich diese Arbeit selbständig verfasst habe, dass ich die verwendeten Quellen und Hilfsmittel vollständig angegeben habe und dass ich die Stellen der Arbeit – einschließlich Tabellen, Karten und Abbildungen –, die anderen Werken oder dem Internet im Wortlaut oder dem Sinn nach entnommen sind, auf jeden Fall unter Angabe der Quelle als Entlehnung kenntlich gemacht habe.  \nWien, 3. Juli 2023     \nAndreas Schmidt  \niii  \nDanksagung  \nIch möchte mich an dieser Stelle bei allen Menschen bedanken, die mich bei der Erstellung dieser Diplomarbeit unterstützt haben.  \nAn erster Stelle möchte ich meinem Betreuer, Herrn Associate Prof. Dipl.-Ing. Dr.techn. Peter Knees, danken, der mein Interesse für dieses Thema geweckt hat und mich richtungsweisend während meiner Arbeit begleitet und mit seinen konskruktiven Ratschlägen unterstützt hat.  \nMein herzlichster Dank gilt meinen Eltern, die mir mein Studium ermöglicht haben. Vielen Dank für eure Unterstützung, eure Geduld und eure Motivation!  \nBesonderer Dank gilt außerdem meiner Schwester Viktoria und meiner Freundin Lydia, sowie meinen Freudinnen und Freunden, die mir stets zur Seite standen.  \niv  \nKurzfassung  \nEine Sammlung von semantischen Musik-Features (d.h. Metriken, die bestimmte Merkmale von Musikstücken quantifizieren), die vom Musik-Streaming-Dienst Spotify angeboten wird, zeigt wissenschaftliche Relevanz im Forschungsfeld Music Information Retrieval aufgrund der Tatsache, dass sie in mehreren vorangegangenen Forschungsarbeiten eingesetzt wurde. Die Algorithmen, die zur Berechnung dieser semantischen Musik-Features verwendet werden sind nicht veröffentlicht und treten daher als „Black-Box“-Algorithmen auf.  \nDiese Arbeit untersucht die Eigenschaften dieser Musik-Features und Ansätze zur Nachbildung der Algorithmen, die zu ihrer Berechnung verwendet werden, bzw. Alternativendazu anhand aktueller Methoden aus dem Bereich Music Information Retrieval. Konkret wurden Algorithmen zur Vorhersage der Features “Danceability”, “Acousticness”,“Instrumentalness”, “Speechiness”, “Liveness”,“Valence”, “Energy” and “Loudness” untersucht. Das Ergebnis dieser Arbeit ist eine Charakterisierung der Eigenschaften der Musik-Features, sowie eine Reihe von Machine-Learning-Modellen, die die Berechnungender Features reproduzieren, und eine Evaluierung, wie gut die Black-Box-Algorithmen reproduziert werden können.  \nv  \nAbstract  \nIn the research field of Music Information Retrieval a set of semantic music features (i.e. descriptors quantifying certain characteristics of pieces of music) that is offered by music streaming service Spotify has gained scientific relevance because of the fact that it has been used in various research works in the past. The algorithms that are originally used for the purpose of calculating these semantic music features are","cbCainujS6kLhhsD","https://ap.wps.com/l/cbCainujS6kLhhsD","pdf",2894785,1,75,"English","en",105,"# Introduction\n## Problem Statement\n## Aim of the Work\n## Methodology\n# Music Information Retrieval\n## Context-Based Music Information Retrieval\n## Content-Based Music Information Retrieval\n# Music Feature Learning using Convolutional Neural Networks\n## Neural Networks\n## Optimization Algorithms\n## Learning Rate\n## Convolutional Neural Networks\n# Experiments\n## Data Set\n## Music Features\n## Experiment Design\n## Experiment Implementation","[{\"question\":\"Why are Spotify’s semantic music features considered scientifically relevant?\",\"answer\":\"They have been used in multiple prior research works within Music Information Retrieval, establishing their scientific value as standardized descriptors.\"},{\"question\":\"What makes the underlying algorithms “black-box” in this thesis?\",\"answer\":\"The algorithms used by Spotify to compute the semantic music features are not publicly available, so only their outputs are observable.\"},{\"question\":\"Which machine learning features and approaches are investigated?\",\"answer\":\"The thesis studies features including Danceability, Acousticness, Instrumentalness, Speechiness, Liveness, Valence, Energy, and Loudness, and evaluates machine learning models that reproduce or approximate the black-box computations.\"}]","Reproduction of Black-Box Music Analysis Algorithms through Machine Learning - Diplomarbeit | PDF",1785809393,189,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"reproduction-of-black-box-music-analysis-algorithms-through-machine-learning-diplomarbeit","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/reproduction-of-black-box-music-analysis-algorithms-through-machine-learning-diplomarbeit/122214/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-04",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why are Spotify’s semantic music features considered scientifically relevant?","Question",{"text":75,"@type":76},"They have been used in multiple prior research works within Music Information Retrieval, establishing their scientific value as standardized descriptors.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What makes the underlying algorithms “black-box” in this thesis?",{"text":80,"@type":76},"The algorithms used by Spotify to compute the semantic music features are not publicly available, so only their outputs are observable.",{"name":82,"@type":73,"acceptedAnswer":83},"Which machine learning features and approaches are investigated?",{"text":84,"@type":76},"The thesis studies features including Danceability, Acousticness, Instrumentalness, Speechiness, Liveness, Valence, Energy, and Loudness, and evaluates machine learning models that reproduce or approximate the black-box computations.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]