Tag Archives: research reproducibility

Journal paper and open dataset for source separation in Orchestra music

As part of the PHENICX project, we have recently published our research results in the task of audio sound source separation, which is the main research topic of one of our PhD students, Marius Miron.

During this work, we developed a method for orchestral music source separation along with a new dataset: the PHENICX-Anechoic dataset. The methods were integrated into the  PHENICX project for tasks as orchestra focus/instrument enhancement. To our knowledge, this is the first time source separation is objectively evaluated in such a complex scenario. 

This is the complete reference to the paper:

M. Miron, J. Carabias-Orti, J. J. Bosch, E. Gómez and J. Janer, “Score-informed source separation for multi-channel orchestral recordings”, Journal of Electrical and Computer Engineering (2016))”

Abstract: This paper proposes a system for score-informed audio source separation for multichannel orchestral recordings. The orchestral music repertoire relies on the existence of scores. Thus, a reliable separation requires a good alignment of the score with the audio of the performance. To that extent, automatic score alignment methods are reliable when allowing a tolerance window around the actual onset and offset. Moreover, several factors increase the difficulty of our task: a high reverberant image, large ensembles having rich polyphony, and a large variety of instruments recorded within a distant-microphone setup. To solve these problems, we design context-specific methods such as the refinement of score-following output in order to obtain a more precise alignment. Moreover, we extend a close-microphone separation framework to deal with the distant-microphone orchestral recordings. Then, we propose the first open evaluation dataset in this musical context, including annotations of the notes played by multiple instruments from an orchestral ensemble. The evaluation aims at analyzing the interactions of important parts of the separation framework on the quality of separation. Results show that we are able to align the original score with the audio of the performance and separate the sources corresponding to the instrument sections.

The PHENICX-Anechoic dataset includes audio and annotations useful for different MIR tasks as score-informed source separation, score following, multi-pitch estimation, transcription or instrument detection, in the context of symphonic music. This dataset is based on the anechoic recordings described in this paper:

Pätynen, J., Pulkki, V., and Lokki, T., “Anechoic recording system for symphony orchestra,” Acta Acustica united with Acustica, vol. 94, nr. 6, pp. 856-865, November/December 2008.

For more information about the dataset and how to download you can access the PHENICX-Anechoic web page.

Leave a comment

Filed under datasets, publications, research

CANTE: Open Algorithm, Code & Data for the Automatic Transcription of Flamenco Singing

Over the last months, several journal publications related to our research on flamenco & technology are finally online.

One of them is a work with my former PhD student, Nadine Kroher (who now moved to Universidad de Sevilla), on the automatic transcription of flamenco singing. Flamenco singing is really challenging in terms of computational modelling, given its ornamented character and variety, and we have designed a system for its automatic transcription, focusing on polyphonic recordings.

flamencoTranscriptionKroherGomez

The proposed system outperforms state of the art singing transcription systems with respect to voicing accuracy, onset detection, and overall performance when evaluated on flamenco singing datasets. We hope it think will be a contribution not only to flamenco research but to other singing styles.

You can read about our algorithm at the paper we published at IEEE TASP, where we present the method, strategies for evaluation and comparison with state of the art approaches. You can not only read, but actually try it, as we published an open source software for the algorithm, plus a music dataset for its comparative evaluation, cante2midi (I will talk about flamenco corpus in another post). All of this to foster research reproducibility and motivate people to work on flamenco music.

¡Olé!

 

 

 

Leave a comment

Filed under publications, research, software