Overview

The 18th International Conference on Music Information Retrieval (ISMIR 2017) will be held in Suzhou, China, from October 23-28, 2017.

Music-Information Retrieval (Music-IR) is a wide-ranging area of study and includes elements from the disciplines of musicology, machine learning, signal processing, cognitive science, library and information science, music therapy, and many others. At ISMIR 2017 we aim to present work covering the state-of-the-art of this entire field of study, allowing researchers from every branch of this field to exchange ideas, compare results, and initiate collaborations.

Accepting submissions for:

  • Conference papers
  • Tutorials
  • Late-breaking papers & demos
  • Musical performances

In addition, ISMIR 2017 will be hosting workshops with their own submission requirements.

Topics of Interest include but are not limited to:

Music-IR data and fundamentals such as:

  • Music signal processing
  • Symbolic music processing
  • Metadata, tags, linked data, and semantic web
  • Lyrics and other textual data, web mining, and natural language processing
  • Multimodal approaches to Music-IR

Extracting musical features and properties such as

  • Genre
  • Mood and emotion
  • Instrumentation
  • Melody and harmony
  • Pitches and chords
  • Key and mode
  • Rhythm, beat, and tempo
  • Structure
  • Timbre
  • Expressive and performative elements

Estimating music metadata such as:

  • Artist
  • Album
  • Release date
  • Popularity
  • Tags
  • Cover song detection

Methods for manipulating, analyzing, and processing musical sequences such as:

  • Sound source separation
  • Alignment, synchronization, and score following
  • Automatic transcription
  • Music synthesis, transformation, and augmentation
  • Fingerprinting

Applications for Music-IR such as:

  • Music-IR for health and therapy
  • Music-IR for learning
  • Music-IR for indexing, querying, and retrieving music
  • Music-IR for pattern matching and detection
  • Music-IR for determining similarity
  • Music-IR for music recommendation
  • Music-IR for music training
  • Music-IR for gaming
  • Music-IR for enhanced performances
  • Music-IR for business and marketing
  • Music-IR for digital libraries and archives

Domain-specific knowledge such as:

  • Computational music theory
  • Computational musicology
  • Cognitive Music-IR
  • Representations of music

Methodological and social concerns such as:

  • Corpus creation
  • Annotation methodology
  • Evaluation methodology
  • Legal issues
  • Ethical issues

User-based concerns such as:

  • User behavior
  • User modeling
  • Human-computer interfaces

Scientific Program Notes

All papers will go through a double-blind review. Each paper will be assigned at least three reviewers per submission. All accepted papers will be presented at ISMIR 2017 as either a poster or an oral presentation. The Program Committee will determine which submissions will be presented orally and which will be presented as posters; this determination will not be based on the relevance or potential impact of the papers, but rather on the content and the best method to reach each paper’s intended audience. Each paper may have a maximum of six pages of scientific content (including figures and possible references) and one additional page which may only contain references.