ISMIR 2020 : International Society for Music Information Retrieval Conference

The annual conference of the International Society for Music Information Retrieval (ISMIR) is the world’s leading research forum on processing, analyzing, searching, organizing and accessing music-related data. The 21st ISMIR conference (https://ismir.github.io/ISMIR2020/), to be held in Montreal, Canada, welcomes contributions related to any aspect of Music IR, including foundations and theories for music processing, evaluation algorithms, applications, and analysis.

Important Dates

Time zone : Anywhere On Earth (AOE)

Abstract Submission, April 10, 2020

Final Submission, April 17, 2020

Notification of Acceptance, June 12, 2020

Camera-Ready Upload, July 6, 2020

There will be no extension to the submission deadlines. However, authors of registered papers on or before April 10th will be allowed to upload new versions of their papers until April 17th.

Submission Guidelines

All submissions must comply with the following requirements:

• Length (6+n pages): papers must contain at most 6 pages of scientific content (including figures and tables), with additional optional pages that contain only references and acknowledgments.

• Templates: papers must be submitted in PDF format using the ISMIR 2020 templates (LaTex or Word). You may not manipulate the style files in any way.

• File size: submitted PDF files must be at most 4MB in size. Please compress images and figures as necessary before submitting.

• Originality: papers must be original contributions. They cannot have been published elsewhere nor currently submitted for publication anywhere else. All relevant work, including direct quotations from your own work, should be cited.

• Submissions:

  • Submission system: We will use Microsoft CMT.

  • Supplementary material: In addition to the PDF file of the manuscript, authors will be able to also upload supplementary files for their submission, such as audio samples, code or additional results. We strongly encourage authors to do this.

  • Main message: During submission, authors will be asked for ONE line of text stating the main take-home message of their work. This information will NOT be available to reviewers or meta-reviewers; it is intended to help organize cohesive sessions for the conference

• Anonymity: ISMIR reviews are double-blind; authors and reviewers will be anonymous to each other. For the initial submission, please keep the author list from appearing in the text. Also, do not add any links/URLs that could reveal authors’ identify (e.g., to github pages). Likewise, make any self-citations in the third person, although self-citations should be kept to a minimum. Note that acknowledgments must not be included in the anonymized submission.

• Preprints: to maintain the legitimacy for our double-blind review process, we strongly discourage authors from posting near duplicate manuscripts on public archives (technical reports, arXiv, etc.). In the same spirit, to protect our double-blind reviewing process, authors need to make sure they do not promote their work in any way during the review process (social media, blog, mailing-list, etc.), since this may prevent preserving anonymity.

Submissions that deviate from any of the above requirements will be automatically rejected.

Review Process

In order to ensure the quality of the papers published at the conference, all submissions will be evaluated according to the following criteria:

• Novelty
• Scholarly/scientific quality
• Reusable insights
• Stimulation potential
• Appropriateness of topic
• Importance
• Readability and paper organization

We especially call for submissions that explicitly discuss reusable insights, that is, insights that may go beyond the scope of the paper, domain or application, to build up consistent knowledge across the MIR community.

The reviewing workflow follows a two-tier model, i.e. with one level of “regular” reviewers, and another level of “meta-reviewers”. All papers are examined by at least three reviewers.

After a long discussion, we decided to refrain from installing a classical rebuttal phase. Instead, we will create a feedback form to allow authors who strongly believe that their submission has been massively misjudged or rejected because of subjective differences in scientific opinion to contact the program chairs, who will then look into the matter closely. Note that this form is not meant to argue against objective complaints on the technical or scientific content of the submission, but only to identify and flag unfair or highly subjective reviews or meta-reviews.

Accepted papers must be presented at the conference by one of the authors, and at least one of the authors must register before the deadline given for author registration (TBD). Failure to register before the deadline will result in automatic withdrawal of the paper from the conference proceedings and program.

Topics of Interest

ISMIR 2020 welcomes full-paper contributions to any aspect of Music IR. Relevant topics include, but are not limited to:

• MIR fundamentals and methodology: music signal processing; symbolic music processing; metadata, tags, linked data, and semantic web; lyrics and other textual data, web mining, and natural language processing; multimodality.

• Domain knowledge: representations of music; music acoustics; computational music theory and musicology; cognitive MIR; machine learning/artificial intelligence for music.

• Musical features and properties: melody and motives; harmony, chords and tonality; rhythm, beat, tempo; structure, segmentation, and form; timbre, instrumentation, and voice; musical style and genre; musical affect, emotion and mood; expression and performative aspects of music.

• MIR tasks: sound source separation; music transcription and annotation; optical music recognition; alignment, synchronization, and score following; music summarization; music synthesis and transformation; fingerprinting; automatic classification; indexing and querying; pattern matching and detection; similarity metrics.

• Evaluation, datasets, and reproducibility: evaluation methodology; evaluation metrics; novel datasets and use cases; annotation protocols; reproducibility; MIR tasks.

• Philosophical and ethical discussions: philosophical and methodological foundations; legal and societal aspects of MIR; ethical issues related to designing and implementing MIR tools and technologies.

• Human-centered MIR: user behavior analysis and mining, user modeling; human-computer interaction and interfaces; personalization; user-centered evaluation.

• Applications: digital libraries and archives; music retrieval systems; music recommendation and playlist generation; music and health, well-being and therapy; music training and education; music composition, performance, and production; gaming; business and marketing. 

For additional information, you can reach the program chairs via email at [email protected].