Keynote Speaker

TM Krishna / Karnatik Musician, Author & Activist
2022-12-06 | 09:00 (Asia/Calcutta)
Indian art music continues to evolve in current performance practice, while staying within the framework provided by some of the immutable axiomatic concepts that define the music culture. The changes that lead the evolution of the music culture are guided by practitioners and influenced by the evolving socio-cultural, socio-political or performance and aesthetic considerations. In this talk, we focus on the evolution of Indian art music from the perspective of performance and aesthetics, highlighting some important milestones around the melodic and rhythmic systems in Indian art music. Focusing on recent developments and our own influences on performance practice and aesthetics, we discuss our effort and approaches to create more inclusive roles in music composition and performance. We further aim to provide concrete examples and formulations of the abstractions in current performance and aesthetics. We propose thoughts and ideas that can help current MIR formulations and solutions to go beyond the limiting assumptions based on current music performance practices and (often rigid) structures, and focus on the music abstractions that are more fundamental to our understanding, appreciation and analysis of Indian art music.
TM Krishna, is one of the pre-eminent vocalists in the rigorous Karnatik tradition of India's classical music. As a public intellectual, Krishna speaks and writes about issues affecting the human condition and about matters cultural. As a vocalist, he has made path-breaking innovations in both the style and substance of his concerts. His award-winning book, A Southern Music – The Karnatik Story, published by Harper Collins in 2013 was a first-of-its-kind philosophical, aesthetic and socio-political exploration of Karnatik music. TM Krishna has partnered with individuals and collectives working at the intersections of social change, a new politics for contemporary India, a fresh new imagining of the wider universe of the Arts. In 2016, TM Krishna received the prestigious Ramon Magsaysay Award in recognition of "his forceful commitment as artist and advocate to art’s power to heal India’s deep social divisions".

Poster Sessions

Paper Session - 3 (Special Call)

Session Chair: Rafael Caro Repetto (University of Music and Performing Arts Graz)

2022-12-06 | 10:00 (Asia/Calcutta)

Browse the active poster session's channels, joining calls to ask questions and discuss research with presenters, and leave comments in the channel for asynchronous chatting later.

  • P3-01*: Raga Classification From Vocal Performances Using Multimodal Analysis
    Martin Clayton, Preeti Rao, Nithya Nadig Shikarpur, Sujoy Roychowdhury, Jin Li
  • P3-02*: Traces of Globalization in Online Music Consumption Patterns and Results of Recommendation Algorithms
    Oleg Lesota, Emilia Parada-Cabaleiro, Elisabeth Lex, Navid Rekabsaz, Stefan Brandl, Markus Schedl
  • P3-03: Network Analyses for Cross-Cultural Music Popularity
    Kongmeng Liew, Vipul Mishra, Yangyang Zhou, Elena V. Epure, Romain Hennequin, Shoko Wakamiya, Eiji Aramaki
  • P3-04: Three related corpora in Middle Byzantine music notation and a preliminary comparative analysis
    Polykarpos Polykarpidis, Dionysios Kalofonos, Dimitrios Balageorgos, Christina Anagnostopoulou
  • P3-05: Playing Technique Detection by Fusing Note Onset Information in Guzheng Performance
    Dichucheng Li, Yulun Wu, Qinyu Li, Jiahao Zhao, Yi Yu, Fan Xia, Wei Li
  • P3-06: KDC: an open corpus for computational research of dastgāhi music
    Babak Nikzat, Rafael Caro Repetto
  • P3-07: Inaccurate Prediction or Genre Evolution? Rethinking Genre Classification
    Ke Nie
  • P3-08: In Search of Sañcāras: Tradition-informed Repeated Melodic Pattern Recognition in Carnatic Music
    Thomas Nuttall, Genís Plaja-Roglans, Lara Pearson, Xavier Serra
  • P3-09: Automatic Chinese National Pentatonic Modes Recognition Using Convolutional Neural Network
    Zhaowen Wang, Mingjin Che, Yue Yang, Wen Wu Meng, Qinyu Li, Fan Xia, Wei Li
  • P3-10: Teach Yourself Georgian Folk Songs Dataset: A Annotated Corpus Of Traditional Vocal Polyphony
    David Gillman, Atalay Kutlay, Uday Goyat
  • P3-11: Adapting meter tracking models to Latin American music
    Lucas S Maia, Martín Rocamora, Luiz W P Biscainho, Magdalena Fuentes
  • P3-12: Critiquing Task- versus Goal-oriented Approaches: A Case for Makam Recognition
    Kaustuv Kanti Ganguli, Sertan Şentürk, Carlos Guedes
  • P3-13: A Dataset for Greek Traditional and Folk Music: Lyra
    Charilaos Papaioannou, Ioannis Valiantzas, Theodore Giannakopoulos, Maximos Kaliakatsos-Papakostas, Alexandros Potamianos
  • P3-14: Analysis and detection of singing techniques in repertoires of J-POP solo singers
    Yuya Yamamoto, Juhan Nam, Hiroko Terasawa

An asterisk (*) indicates long presentations (paper award candidates)

Paper Session - 4

Session Chair: Vinoo Alluri (IIIT Hyderabad)

2022-12-06 | 13:30 (Asia/Calcutta)

Browse the active poster session's channels, joining calls to ask questions and discuss research with presenters, and leave comments in the channel for asynchronous chatting later.

  • P4-01*: Performance MIDI-to-score conversion by neural beat tracking
    Lele Liu, Qiuqiang Kong, Veronica Morfi, Emmanouil Benetos
  • P4-02: Symbolic Music Loop Generation with Neural Discrete Representations
    Sangjun Han, Hyeongrae Ihm, Moontae Lee, Woohyung Lim
  • P4-03: Automatic music mixing with deep learning and out-of-domain data
    Marco A Martinez Ramirez, Weihsiang Liao, Chihiro Nagashima, Giorgio Fabbro, Stefan Uhlich, Yuki Mitsufuji
  • P4-04: Music-STAR: a Style Translation system for Audio-based Re-instrumentation
    Mahshid Alinoori, Vassilios Tzerpos
  • P4-05: Learning Unsupervised Hierarchies of Audio Concepts
    Darius Afchar, Romain Hennequin, Vincent Guigue
  • P4-06: Multi-objective Hyper-parameter Optimization of Behavioral Song Embeddings
    Massimo Quadrana, Antoine Larreche-Mouly, Matthias Mauch
  • P4-07: ATEPP: A Dataset of Automatically Transcribed Expressive Piano Performance
    Huan Zhang, Jingjing Tang, Syed Rm Rafee, Simon Dixon, George Fazekas, Geraint A. Wiggins
  • P4-08: PDAugment: Data Augmentation by Pitch and Duration Adjustments for Automatic Lyrics Transcription
    Chen Zhang, Jiaxing Yu, Luchin Chang, Xu Tan, Jiawei Chen, Tao Qin, Kejun Zhang
  • P4-09: Parameter Sensitivity of Deep-Feature based Evaluation Metrics for Audio Textures
    Chitralekha Gupta, Yize Wei, Zequn Gong, Purnima Kamath, Zhuoyao Li, Lonce Wyse
  • P4-10: Stability of Symbolic Feature Group Importance in the Context of Multi-Modal Music Classification
    Igor Vatolkin, Cory Mckay
  • P4-11: Multi-pitch Estimation meets Microphone Mismatch: Applicability of Domain Adaptation
    Franca Bittner, Marcel Gonzalez, Maike L Richter, Hanna Lukashevich, Jakob Abeßer
  • P4-12: Melody transcription via generative pre-training
    Chris Donahue, John Thickstun, Percy Liang
  • P4-13: Source Separation of Piano Concertos with Test-Time Adaptation
    Yigitcan Özer, Meinard Müller
  • P4-14: Counterpoint Error-Detection Tools for Optical Music Recognition of Renaissance Polyphonic Music
    Martha E Thomae Elias, Julie Cumming, Ichiro Fujinaga
  • P4-15: A Dataset of Symbolic Texture Annotations in Mozart Piano Sonatas
    Louis Couturier, Louis Bigo, Florence Leve
  • P4-16: Violin Etudes: A Comprehensive Dataset for f0 Estimation and Performance Analysis
    Nazif Can Tamer, Pedro Ramoneda, Xavier Serra
  • P4-17: Checklist Models for Improved Output Fluency in Piano Fingering Prediction
    Nikita Srivatsan, Taylor Berg-Kirkpatrick

An asterisk (*) indicates long presentations (paper award candidates)

Special Sessions

Special Session - 1: Enhancing music listening with MIR

Moderator: Xavier Serra (Universitat Pompeu Fabra)

2022-12-06 | 16:00 (Asia/Calcutta)

In this panel we will discuss the research challenges and opportunities related to the development of new MIR technologies and services to support music listening.

Panelists: Anna Gatzioura (Chordify), Fabien Gouyon (Pandora), Thomas Lidy (Utopia), Hugo Rodrigues (

Music Sessions

Social Events

ISMIR Music Concert

Hindustani vocals: Kaustuv Kanti Ganguli

Carnatic vocals: Vignesh Ishwar

Harmonium: Ravindra Katoti

Carnatic violin: Sayee Rakshith

Tabla: Tejovrush Joshi

Mridangam: Sumesh Narayanan

2022-12-06 | 18:30 (Asia/Calcutta)

ISMIR 2022 Music Concert will feature a Jugalbandi vocal Indian art music concert by Kaustuv Kanti Ganguli and Vignesh Ishwar. The jugalbandi concert will aim to showcase the commonalities, differences and nuances of Hindustani and Carnatic music, the two predominant art music traditions of India. Kaustuv Kanti Ganguli will be accompanied by Ravindra Katoti on the harmonium and Tejovrush Joshi on the tabla. Vignesh Ishwar will be accompanied by Sayee Rakshith on the violin and Sumesh Narayanan on the mridangam. Kaustuv and Vignesh are seasoned professional musicians and MIR researchers who can bring their expertise and understanding to put together an enthralling performance interesting to the conference participants. A brief biography of the artists and additional details of the concert will be available soon.

Virtual Special Sessions

Special Session B (Online): PhD in MIR - Challenges and Opportunities

Moderator: Meinard Müller (International Audio Laboratories Erlangen)

2022-12-06 | 22:00 (Asia/Calcutta)

Music information retrieval (MIR) is an exciting research field related to different disciplines, including signal processing, machine learning, information retrieval, psychology, musicology, and the digital humanities. This diversity opens up many opportunities for challenging, interdisciplinary, and fascinating research projects at the intersection of engineering and humanities. However, younger researchers can also feel overwhelmed by the variety and complexity of MIR research questions. In this session, we will have an informal exchange of ideas and experiences, inviting doctoral candidates and more experienced MIR researchers. Responding to questions from the audience, we hope this interactive session will be helpful for current PhD students and students considering a PhD in MIR.