Impressions from ICASSP 2017

The signal processing community is very into machine learning. Although I am not sure of the implications of this fact, this intersection already produced very interesting results – such as Smaragdis et al.’s work. Lots of papers related to deep learning were presented. Although in many cases people were naively applying DNN or LSTMs to a new problem, there also was (of course) amazing work with inspiring ideas – I highlight some:

  • Koizumi et al. propose using reinforcement learning for source separation. This work introduces how to use reinforcement learning for audio signal processing.
  • Ewert et al. propose using a variant of dropout that can be used to induce models to learn specific structures by using information from weak labels.
  • Ting-Wei et al. propose doing frame-level predictions with a fully convolutional model that also uses gaussian kernel filters (first introduced by them) trained with clip-level annotations in a weakly-supervised learning setup.

Continue reading

Slides: Deep learning for music data processing – a personal (re)view

I was invited to give a talk to the Deep Learning for Speech and Language Winter Seminar @ UPC,  Barcelona. Since UPC is the university where I did my undergaduate sudies, it was a great pleasure to give an introductory talk about how our community is using deep learning for approaching music technology problems.

Download the slides!

Overall, the talk was centered in reviewing the state-of-the-art (1988-2016) in deep learning for music data processing in order to boost some discussion about current trends. Several key papers were chronologically listed and briefly described: pioneer papers using MLP [1], RNNs [2], LSTMs [3] and CNNs [4] for music data processing; and pioner papers using symbolic data [1], spectrograms [5] and waveforms [6] – among others.

Continue reading

Lack of annotated music data? Restrict the solution space!

Given that several relevant researchers in our field were in Barcelona for being part of the jury of Ajay‘s and Sankalp‘s PhD thesis defense, the MTG hosted a very interesting seminar. Among other topics, the potential impact of deep learning in our field was discussed and almost everyone agreed that it seems that end-to-end learning approaches are not successful because no large-scale (annotated) music collections are available for research benchmarking. And indeed, most successful deep learning approaches use those models as mere feature extractors or as hierarchical classifiers build on top of hand-crafted features.


Continue reading

Rationalizing the design of deep learning models for music signals

A brief review of the state-of-the-art in music informatics research (MIR) and deep learning reveals that such models achieved competitive results in a relatively short amount of time – most relevant papers were published during the last 5 years. Many researchers successfully used deep learning for several tasks: onset detection, genre classification, chord estimation, auto-tagging or source separation. Even some researchers declare that is the time for a paradigm shift: from hand-crafted features and shallow classifiers to deep processing models. In fact, in the past, introducing machine learning for global modeling (ie. classification) resulted in a significant state-of-the-art advance – no one doubts about that. And now, some researchers think that another advance could be done by using data-driven feature extractors instead of hand-crafted features – meaning that these researchers propose to fully substitute the current pipeline by machine learning. However, deep learning for MIR is still in its early ages. Current systems are based on solutions proposed in the computer vision, natural language or speech research fields. Therefore, now it is time to understand and adapt these for the music case.

Continue reading