Last week saw the 16th International Society for Music Information Retrieval Conference (ISMIR 2015), which took place in Malaga, Spain. ISMIR is the leading conference in the field of music informatics, and is currently the top ranking publication in terms of h-index for Music & Musicology (source: Google Scholar).
Queen Mary University of London, and more specifically its Centre for Digital Music (C4DM), had as in previous years a strong presence in the conference, both in terms of numbers and overall impact.
On Tutorials, the following were co-organised by C4DM members:
- Why singing is interesting (Simon Dixon, Masataka Goto, Matthias Mauch). Slides (PDF)
- Automatic music transcription (Zhiyao Duan, Emmanouil Benetos). Slides & Material
On Keynotes, Prof. Mark Sandler gave a talk on "Integrating music information sources for music production and consumption".
In the Technical Programme, the following papers were authored/coauthored by C4DM members:
- Audio Chord Recognition with a Hybrid Recurrent Neural Network (Siddharth Sigtia, Nicolas Boulanger-Lewandowski, Simon Dixon)
- Analysis of Intonation Trajectories in Solo Singing (Jiajie Dai, Matthias Mauch, Simon Dixon)
- Combining Features for Cover Song Identification (Julien Osmalskyj, Peter Foster, Simon Dixon, Jean-Jacques Embrechts)
- An Efficient Temporally-Constrained Probabilistic Model for Multiple-Instrument Music Transcription (Emmanouil Benetos, Tillman Weyde)
For the MIREX 2015 public evaluation session, the following submissions were made:
- Chris Cannam, Emmanouil Benetos, Matthias Mauch, Matthew E. P. Davies, Simon Dixon, Christian Landone, Katy Noland, and Dan Stowell, "MIREX 2015: VAMP Plugins from the Centre for Digital Music". Participated in the following tasks: Music Structure Segmentation, Audio Onset Detection, Audio Tempo Estimation, Audio Key Detection, Audio Beat Tracking, Audio Chord Estimation, Multiple Fundamental Frequency Estimation & Note Tracking.
- Emmanouil Benetos and Tillman Weyde, "Multiple-F0 estimation and note tracking for MIREX 2015 using a sound state-based spectrogram factorization model". Participated in the Multiple Fundamental Frequency Estimation & Note Tracking task.
In addition, C4DM co-organised the MIREX 2015 task on Music/Speech Classification and Detection (task captains: Tillman Weyde, Aggelos Pikrakis, Emmanouil Benetos, Daniel Wolff, Dan Tidhar)
At the Business Meeting, Dr Simon Dixon as President of ISMIR presented the society's work and activities to the community.
For the Demos & Late-Breaking News session, the following submissions were authored/coauthored by C4DM members:
- Faults in the Latin Music Database and with its Use (Bob L. Sturm)
- Estimation of the Reliability of Multiple Rhythm Features Extraction from a Single Descriptor (Elio Quinton, Mark Sandler, Simon Dixon)
- Spatially Rendering Decomposed Recordings - Integrating Score-Informed Source Separation and Semantic Playback Technologies (Florian Thalmann, Sebastian Ewert, Mark Sandler, Geraint A. Wiggins)
- Folk Music Style Modelling by Recurrent Neural Networks with Long Short Term Memory Units (Bob Sturm, João Felipe Santos and Iryna Korshunova)
- Re-evaluating the Scattering Transform (Francisco Rodríguez-Algarra and Bob L. Sturm)
- The Organ Web App (Beici Liang, George Fazekas, Mark Sandler)
- GeoLocation-Adaptive Music Player (Alfonso Perez-Carrillo, Florian Thalmann, Thomas Wilmering and Mark Sandler)
- Augmented Live Coding: Harnessing Linked Data in Musical Performances (Alo Allik)
- Auralisation of Deep Convolutional Neural Networks: Listening to Learned Features (Keunwoo Choi, Jeonghee Kim, George Fazekas and Mark Sandler)
- Navigating Ontological Structures based on Feature Metadata Using the Semantic Music Player (Florian Thalmann, Alfonso Perez Carrillo, György Fazekas, Geraint Wiggins and Mark Sandler)
- An Ontology for Abstract, Hierarchical Music Representation (Nicholas Harley and Geraint Wiggins)
- The Digital Music Lab: A Big Data Infrastructure for Digital Musicology (Tillman Weyde, Stephen Cottrell, Jason Dykes, Emmanouil Benetos, Daniel Wolff, Alexander Kachkaev, Dan Tidhar, Mark Plumbley, Simon Dixon, Stephen Hargreaves, Mathieu Barthet, Nicolas Gold, Samer Abdallah, Dan Tidhar, Mark Plumbley)
- Vocal Imitation of Pitch, Spectral Shape and Loudness Envelopes (Adib Mehrabi, Simon Dixon and Mark Sandler)
In the Unconference session, C4DM members co-organised a discussion on Big Data for Musicology (Tillman Weyde, Emmanouil Benetos, Daniel Wolff)
Finally, on the organisational side, the following C4DM members were members of the programme committee: Emmanouil Benetos, Simon Dixon, Sebastian Ewert, George Fazekas, Dorien Herremans, Matthias Mauch; and Bob Sturm chaired the oral session on Music Similarity.
(Author: Emmanouil Benetos - 2nd November 2015)