ZemPod: A semantic web approach to podcasting Journal Of Web Semantics 2008 Oscar Celma, Music Technology Group, Spain Yves Raimond, Centre for Digital.

Slides:



Advertisements
Similar presentations
Presented to the ALCTS FRBR Interest Group, ALA Annual, 24 June 2011
Advertisements

Connecting Social Content Services using FOAF, RDF and REST Leigh Dodds, Engineering Manager, Ingenta Amsterdam, May 2005.
Pseudo-Relevance Feedback For Multimedia Retrieval By Rong Yan, Alexander G. and Rong Jin Mwangi S. Kariuki
Presentation by Priyanka Sawarkar
A Stepwise Modeling Approach for Individual Media Semantics Annett Mitschick, Klaus Meißner TU Dresden, Department of Computer Science, Multimedia Technology.
Podcasting. What is Podcasting? A collection of technologies for distributing Audio and Video over the Internet Distributed by a RSS (Really Simple Syndication)
Semantic Web 2 06 T 0006 Yoshiyuki Osawa. Aim of Semantic Web Information which users needs is collected by using a computer. Information on the web is.
Future Software Architectures Combining the Web 2.0 with the Semantic Web to realize future Web Communities Maarten Visser
THE UNIVERSITY OF HONG KONG WEB BY DANIEL CHURCHILL 2.0.
Semantic Web and Web Mining: Networking with Industry and Academia İsmail Hakkı Toroslu IST EVENT 2006.
ADVISE: Advanced Digital Video Information Segmentation Engine
Presentation Outline  Project Aims  Introduction of Digital Video Library  Introduction of Our Work  Considerations and Approach  Design and Implementation.
1 MPEG-21 : Goals and Achievements Ian Burnett, Rik Van de Walle, Keith Hill, Jan Bormans and Fernando Pereira IEEE Multimedia, October-November 2003.
Architecture & Data Management of XML-Based Digital Video Library System Jacky C.K. Ma Michael R. Lyu.
CSC 101 Slide Show Ashley Carroll. Podcast What is Podcasting? Podcasting is the distribution of audio or video files, such as radio programs or music.
ReQuest (Validating Semantic Searches) Norman Piedade de Noronha 16 th July, 2004.
Visual Information Retrieval Chapter 1 Introduction Alberto Del Bimbo Dipartimento di Sistemi e Informatica Universita di Firenze Firenze, Italy.
: is a web site, usually maintained by an individual with regular entries of commentary, descriptions of events, or other material such as graphics or.
Jesse Embley CSC 101 Spring 2006 Podcast, Blog, Wiki, RSS.
Metadata Presentation by Rick Pitchford Chief Engineer, School of Communication COM 633, Content Analysis Methods Fall 2009.
CSC By: Shawn Desmond Podcasts, Blogs, Wiki, RSS.
Web 2.0: Concepts and Applications 3 Syndicating Content.
Lecture-8/ T. Nouf Almujally
RESTful Publish Subscribe Xiang Su
Mr. Ulmer Multimedia Technology.  A method of distributing multimedia files (such as audio programs and music videos) over the Internet, using either.
Podcasting 101..and more. Workshop Objectives: Introduce iTunes: abundance of resources, multi-media organizer, classroom tool You do not need an iPod.
Web 2.0: Concepts and Applications 3 Syndicating Content.
Consider ways to use social software in your professional learning and school.
Practical RDF Chapter 1. RDF: An Introduction
Adaptive News Access Daniel Billsus Presented by Chirayu Wongchokprasitti.
RDA data and applications Gordon Dunsire Presented to staff of the British Library, Boston Spa, 20 Mar 2014.
Exploiting Ontologies for Automatic Image Annotation M. Srikanth, J. Varner, M. Bowden, D. Moldovan Language Computer Corporation
Podcasts Spring, 2008 By Linda Kenney Modified Feb. 6, 2008.
Semantic Web Applications GoodRelations BBC Artists BBC World Cup 2010 Website Emma Nherera.
Of 33 lecture 10: ontology – evolution. of 33 ece 720, winter ‘122 ontology evolution introduction - ontologies enable knowledge to be made explicit and.
 Copyright 2008 Digital Enterprise Research Institute. All rights reserved. Semantic on the Social Semantic Desktop.
The Future of Cataloging Codes and Systems: IME ICC, FRBR, and RDA by Dr. Barbara B. Tillett Chief, Cataloging Policy & Support Office Library of Congress.
5 - 1 Copyright © 2006, The McGraw-Hill Companies, Inc. All rights reserved.
Podcasts/Podcasting Podcasting is the downloading of audio broadcasts to your computer. Podcasting entails audio content that is delivered via an RSS.
RDFa, Microformats, and Atom Semantic Web Presented by: Anuradha Kandula Instructor: Steven Seida.
Resource Description and Access Deirdre Kiorgaard ACOC Seminar, September 2007.
Lifecycle Metadata for Digital Objects November 1, 2004 Descriptive Metadata: “Modeling the World”
Using Several Ontologies for Describing Audio-Visual Documents: A Case Study in the Medical Domain Sunday 29 th of May, 2005 Antoine Isaac 1 & Raphaël.
Evidence from Metadata INST 734 Doug Oard Module 8.
RDA DAY 1 – part 2 web version 1. 2 When you catalog a “book” in hand: You are working with a FRBR Group 1 Item The bibliographic record you create will.
Personalized Interaction With Semantic Information Portals Eric Schwarzkopf DFKI
Intellectual Works and their Manifestations Representation of Information Objects IR Systems & Information objects Spring January, 2006 Bharat.
Of 33 lecture 1: introduction. of 33 the semantic web vision today’s web (1) web content – for human consumption (no structural information) people search.
Information Retrieval
What the Principal Needs to Know About Web 2.0 by Rita Lewis Smith October 19, 2010.
Digital Video Library Network Supervisor: Prof. Michael Lyu Student: Ma Chak Kei, Jacky.
RDA: a new cataloging standard for a digital future RDA Update Forum ALA Midwinter Meeting Philadelphia, PA January 13, 2008 John Attig ALA Representative.
Beginning Podcasting November 5 th and 17 th 4 p.m. to 7 p.m.
MPEG-7 Audio Overview Ichiro Fujinaga MUMT 611 McGill University.
Content-Based MP3 Information Retrieval Chueh-Chih Liu Department of Accounting Information Systems Chihlee Institute of Technology 2005/06/16.
1 A Medical Information Management System Using the Semantic Web Technology Networked Computing and Advanced INFORMATION MANAGEMENT, NCM '08. Fourth.
Podcasts. (derived from Apple's "iPod" and "broadcasting“) a method of publishing audio files to the internet, allowing users to subscribe to a feed and.
Statistical techniques for video analysis and searching chapter Anton Korotygin.
Three Internet Medias Podcast, Blogs, Wiki Jasmine Sampson CSC101.
CASEY A. MULLIN WITH: LALA HAJIBAYOVA SCOTT MCCAULAY DECEMBER 8, 2008 FRBR in RDF: a proof-of-concept model 1 ©2008 Casey A. Mullin.
Subjects in the FR family
Visual Information Retrieval
Feed: RSS/ATOM, Podcast
Introduction Multimedia initial focus
“Real Simple Syndication” (RSS)
Lifecycle Metadata for Digital Objects
ece 627 intelligent web: ontology and beyond
Lesson 9: GUI HTML Editors and Mobile Web Sites
MUMT611: Music Information Acquisition, Preservation, and Retrieval
FRBR and FRAD as Implemented in RDA
Presentation transcript:

ZemPod: A semantic web approach to podcasting Journal Of Web Semantics 2008 Oscar Celma, Music Technology Group, Spain Yves Raimond, Centre for Digital Music, UK August 31 th, 2009

Contents  Introduction  Background  System architecture  Usage scenario  Conclusions 2

Introduction [1/2]  Podcast  Portmanteau of the “iPod” and “broadcast”  A media file distributed in Internet  Use syndication feeds  Explosion in popularity of mobile devices  Make syndication model more attractive  Thousands of audio podcasts are available on the net 3

Introduction [2/2]  There are some limitations of podcasting  No formal description  Only textual description available in HTML  No information about the contents of a podcast session  Consists of a single audio file  Difficult to seek into one of the music tracks  To overcome these limitations  Using traditional audio signal processing  Speech/audio segmentation  Audio identification  Adding semantics to the podcast 4

Contents  Introduction  Background  Multimedia web syndication  Speech/music segmentation  Audio identification  The music ontology  System architecture  Usage scenario  Conclusions 5

Multimedia web syndication [1/2]  File format used for syndication  RSS  Really Simple Syndication (RSS 2.0)  Rich Site Summary (RSS 0.91 and 1.0)  RDF Site Summary (RSS 1.0)  Atom  To standardize feeds notation and autodiscovery  Due to some limitations and incompatibility versions of the RSS family 6

Multimedia web syndication [2/2]  Example of RSS 7

Feeds and the semantic web  Atom/Owl  Aims at capturing the semantics of the Atom syndication format  Feed  Attached metadata  Entry  Holds a text content 8

Speech/music segmentation  Discriminating between speech (or spoken content) versus music  Achieving an automatically and meaningful segmentation of a podcast session  Speech/music segmentation methods  Gaussian Mixture Models (GMM)  Support Vector Machines (SVM) classifiers  Combination of standard Hidden Markov Models and Multilayer Perceptrons 9

Audio identification  Allows identification of unknown music  Audio fingerprint  A unique, compact code derived from perceptually relevant aspects of a recording  Usages  Identification  Authentication  Content-based key generation  Content-based audio retrieval and processing  Hidden Markov Models (HMM)  Can precisely model temporal evolution of audio signals 10

Music ontology [1/2]  Create a formal framework  Describing music-related information  Covering complex editorial information  External Ontologies used by Music Ontology  OWL-Time ontology  Describing the temporal content of Web  Interval, Instant  FRBR  Functional Requirements for Bibliographic Records  Work, Expression, Manifestation, Item  FOAF  Friend Of A Friend  Person, Group, Organization 11

Music ontology [2/2]  Describing a music production workflow 12

Contents  Introduction  Background  System architecture  RDFizing a podcast session  Access and workflow  Awareness of feeds  Resource identifiers  Usage scenario  Conclusions 13

System architecture  Main goal is  Analysing and decomposing a given podcast audio file  RDFizing the podcast information 14

 The system segments the audio file into speech and music sections 15

 Apply speech recognition to extract a list of textual terms 16

 Weight terms’ relevance according to a dictionary of musical terms 17

 Recognize music chunks using fingerprinting 18

 Query a metadata repository to get basic information with the track 19

RDFizing a podcast session  To describe the semantics of a podcast  Using Atom-OWL and music ontology  “From 0 to 2 min, there is someone speaking about Michel Jackson, then there is a recording of a ‘Billie Jean’ in 1983”  Using 2 sub concept of the Event  MusicSegment  Temporal region holding music  SpeechSegment  Temporal region holding speech 20

Access and workflow  REST interface  Representational state transfer  Style of software architecture for distributed hypermedia systems such as WWW  Allow us to access the podcast service   Considering the podcast service is available 21

Access and workflow - Awareness of feeds  Internal representation of this feed  Music ontology/AtomOWL  Can be queried through SPARQL 22 USERhttp://zempod.net/feed POST 201 (Created) Location Identifier

Access and workflow - Resource identifiers  MO/AtomOWL are designed as a hierarchical URI space  Feed  Supports a syndication   Entry  Holds a text content   Item  Actual contents  tem{ITEMID} tem{ITEMID} 23

Contents  Introduction  Background  System architecture  Usage scenario  Submission of the original feed  Analysis of the new entries  Semantic description of the new entries  Conclusions 24

Submission of the original feed 25 ser/billy2rivers/mrss POST 201 (Created) Location Identifier Original feed

Analysis of the new entries  Processing a new podcast session 26

Semantic description of the new entries 27 USER GET

Conclusions  To solve limitations of podcasting  No formal description of a podcast  Difficult to seek into one of the music tracks  Using traditional audio signal processing  Speech/music segmentation  Audio identification  Using semantic web techniques  Transform the current RSS to the Atom/OWL  It will ease some important music information retrieval tasks 28

Related Ontology – MO/Event  To express the production process of a pie ce of music  The main sub-classes of event  Performance, Recording, Arrangement, Composition 29

Related Ontology - FRBR  Functional Requirements for Bibliographic Records  서지 레코드의 기능상 요건  목록규칙이나 목록의 완성을 의도하는 개체 - 관계 모델  서지 레코드의 구조와 관계  목록규칙 제정과 시스템 디자인을 위한 정확한 어휘 제공 30 WORKEXPRESSIONMANIFESTATIONITEM 저작표현형구현형개별자료 is realizedis embodiedis exemplified 실현되다구현되다사례가 되다 is ownedis producedis createdhas a subject 소장되다제작되다창작되다주제로 하다

FRBR – Entities and Relationships (1)  Entities and Primary Relationships 31

FRBR – Entities and Relationships (2)  Entities and “Responsibility” Relationships 32

FRBR – Entities and Relationships (3)  Entities and “Subject” Relationships 33 WORK MENIFESTATIO N CORPORATE BODY PERSON ITEM EXPRESSIONWORK EVENTPLACE OBJECTCONCEPT has as subject

MusicBrainz 34