Spectrogram of swelling trumpet sound
Art+Logic’s Incubator project has made a lot of progress. In a previous post I mentioned that Dr. Scott Hawley’s technique to classify audio involved converting audio to an image and using a Convolutional Neural Network (CNN) to classify the audio based on this image. That image is a spectrogram. I’m going to go into some detail about what we do to create one, and why to the best of my ability.
Next week (18-20 November) I’ll be attending the annual Audio Developer Conference in London. On Tuesday November 19th at 16:00 I’ll be part of a team providing the first public details about the forthcoming MIDI 2.0 standard.
The ADC is usually live-streamed on YouTube as it happens, an unfortunate series of events have endangered that this year — you can learn more about that and consider contributing to the fund that will pay for the recording and livestreaming of conference sessions—I frequently return to the archived videos and point other developers to them for reference.
Check the JUCE YouTube channel for the streams during the event (and come back later for archived recordings, or watch sessions from earlier years).
The full schedule for the event is here.
If you’re attending the event, please do track me down and say ‘hey’.
friz—a Flexible Animation Controller for JUCE
As is often the case, I found myself working on a personal project and had some UI elements that really wanted to have some life to them on the screen.
I tucked into the JUCE documentation expecting to see something that I could use to easily add some personality to the interface, and…didn’t find what I was looking for. There’s a
ComponentAnimator class that supports moving a single component from one set of coordinates/bounds to another linearly, and also modify the component’s alpha value to have it fade in or out.
I was looking for something…more expressive.
SXSW is right around the corner (the Interactive Festival is from March 8-12), and we’ll be there. Mark your calendars and come by and meet us. If you can’t make it to one of the events below, send us a note.
Art+Logic has kicked-off its first software Incubator project, and I was selected to handle the development effort. After meeting Dr. Scott Hawley and being briefed on the technique he uses for classification of audio files using neural networks (NN), and determining current and future features, we were ready to begin the project. While we go through this process, I’ll be documenting it on this blog. (more…)