Code

Sound Event Detection (DCASE 2016 - Task 2 Baseline)

Matlab code for sound event detection using an non-negative matrix factorization with beta-divergence, developed for the 2016 IEEE Challege on Detection and Classification of Acoustic Scenes and Events (DCASE 2016). System trained on 11 classes of office sounds from the DCASE 2016 Task 2 training set. Developed by Emmanouil Benetos.

CODE

Please cite: IEEE Challenge on Detection and Classification of Acoustic Scenes and Events 2016, Task 2 (Synthetic audio sound event detection). http://www.cs.tut.fi/sgn/arg/dcase2016/task-sound-event-detection-in-synthetic-audio

Sound Event Detection (efficient temporally-constrained version)

Matlab code for sound event detection using an efficient temporally-constrained model, based on probabilistic latent component analysis. System trained on office sounds from the DCASE 2013 Office Synthetic challenge. Developed by Emmanouil Benetos.

CODE

Related publication: E. Benetos, G. Lafay, M. Lagrange, and M. D. Plumbley, "Detection of overlapping acoustic events using a temporally-constrained probabilistic model", IEEE International Conference on Acoustics, Speech, and Signal Processing, March 2016.

Automatic Music Transcription (efficient temporally-constrained version)

Matlab code for automatic music transcription using an efficient temporally-constrained multiple-instrument model, based on probabilistic latent component analysis. Developed by Emmanouil Benetos.

CODE

Related publication: E. Benetos and T. Weyde, "An efficient temporally-constrained probabilistic model for multiple-instrument music transcription", in Proc. 16th International Society for Music Information Retrieval Conference, pp. 701-707, Oct. 2015.

Automatic transcription of Turkish makam music

This automatic music transcription method takes as input a .wav recording of instrumental or vocal Turkish makam music with its corresponding makam and returns a list of detected notes (onset, duration, and pitch value in cent scale, centered around a tonic frequency). Developed by Emmanouil Benetos and Andre Holzapfel.

CODE

Related publication: E. Benetos and A. Holzapfel, "Automatic transcription of Turkish microtonal music", Journal of the Acoustical Society of America, vol. 138, no. 4, pp. 2118-2130, Oct. 2015.

Silvet Note Transcription - VAMP Plugin

Silvet, or Shift-Invariant Latent Variable Transcription, is a plugin for polyphonic music transcription, that works with a VAMP plugin host like Sonic Visualiser. The Silvet plugin code was adapted by Chris Cannam from research and a Matlab implementation by Emmanouil Benetos.

CODE AND BINARIES

Related publication: E. Benetos and S. Dixon, "A shift-invariant latent variable model for automatic music transcription", Computer Music J., vol. 36, no. 4, pp. 81-94, Winter 2012.

Automatic Transcription of Pitched and Unpitched Sounds

Matlab code for automatic transcription of music recordings containing pitched sounds and percussion. Method based on probabilistic latent component analysis. Developed by Emmanouil Benetos.

CODE

Related publication: E. Benetos, S. Ewert, and T. Weyde, "Automatic transcription of pitched and unpitched sounds from polyphonic music", in Proc. IEEE International Conference on Acoustics, Speech, and Signal Processing, pp. 3131-3135, May 2014.

Automatic Music Transcription (efficient SI-PLCA version)

Matlab code for automatic music transcription using an efficient version of the Multi-Source Shift-Invariant Probabilistic Latent Component Analysis model. Code supports optional GPU processing. Developed by Emmanouil Benetos.

CODE

Related publication: E. Benetos, S. Cherla, and T. Weyde, "An efficient shift-invariant model for polyphonic music transcription", 6th Int. Workshop on Machine Learning and Music, 4 pages, Sep. 2013.

Sound Event Detection (DCASE 2013 Baseline)

Matlab code for baseline system on acoustic event detection, developed as part of the IEEE D-CASE Challenge. Developed by Dimitrios Giannoulis and Emmanouil Benetos.

CODE

Related publication: D. Giannoulis, D. Stowell, E. Benetos, M. Rossignol, M, Lagrange, and M.D. Plumbley, "A database and challenge for acoustic scene classification and event detection", in Proc. 21st European Signal Processing Conf., 5 pages, Sep. 2013.

Score-informed Piano Transcription

Matlab code for score-informed piano transcription. Particularly useful for automated piano tutoring systems. Developed by Emmanouil Benetos.

CODE     DATASET

Related publication: E. Benetos, A. Klapuri, and S. Dixon, "Score-informed transcription for automatic piano tutoring", in Proc. 20th European Signal Processing Conf., pp. 2153-2157, Aug. 2012.