GYM Electro Groove Generator Creates Workout Music On The Fly

YouTube Preview Image

When are algorithms better than musicians?

That’s the question raised by Gym Electro Groove Generator, a new iOS application that generates workout music on the fly.

Here’s what the developer has to say about it:

HI. We created this app for working out at the gym. We found that it not only was it difficult to get the right playlist to help you exercise, but often tracks were too slow or not the right mood.

GYM e.g.g. generates electro grooves on the fly and constantly evolves them. It uses rules for generating patterns and note scales, but there is a certain amount of randomness, so you shouldnt hear the same track twice. Each track blends in to the next one.

You can tap the beat you want to match your preferred tempo. There is an ‘unlike’ button for the occasions that it gets messy or not to your liking.

As the video above demonstrates, the results are function, but not inspired. For cooling off, you might want to switch to Eno’s new generative album, Scape.

GYM e.g.g. is available now in the App Store for $1.99.

Think there’s a time and place for music generated by algorithms? Let us know what you think!

Note: The developer followed up to say:

Hello. I sent you a post today about GYM e.g.g.. I forgot to add that it really is not supposed to generate music to just listen to but it is only background music with sharp pronounced beats and grooves purely to exercise too

I understand it does not create complete complex songs to just listen and enjoy. It is designed to keep you in sync while exercising. I wouldn’t want to give a false impression that this generates chart quality music on the fly. I hope that makes sense.

via Rob W


12 thoughts on “GYM Electro Groove Generator Creates Workout Music On The Fly

  1. It’s not a bad idea to just generate the music that you anyway *don’t* really listen to!

    Hard to judge this app though, when they never let it go for more than 6-7 seconds (well, we *can* judge that it sounds like a very tired DJ noodling some later Kraftwerk samples).

    It always seems that algorithmic songs *should* be doable, since we can find so many similarities within all music genres.
    But every algorithmic imitation piece I’ve heard fails wrt its style – result is too static or it loses cohesion (or both).

    The ears/music-part of brain is excellent at finding similarities at several levels AND also at getting annoyed with too UN-similar sounds such as created by randomness.
    Still too early to bash algorithms in general though, most don’t yet try to model rules at any levels above basic melodic and rhythmic variations.

    Well-loved. Like or Dislike: Thumb up 6 Thumb down 0

    • I like the idea of algorithmic variations, but not somuch when the software starts doing everything.

      People also underestimate how much thought goes into the sound of dance music. The melody and chords may be formulaic, but usually you can identify a song in a fraction of a second because of its unique sound.

      Like or Dislike: Thumb up 2 Thumb down 3

  2. If you can’t come up with better workout music than this, you deserve to have to listen to it. If you have an iAnything, you have iTunes, just for starters. Don’t play Kraftwerk samples, play some real Kraftwerk. You can even play religious or political podcasts and work up a head of steam for a run. :P

    Like or Dislike: Thumb up 1 Thumb down 1

  3. The point of motivational music is always to push the listener, not change in downward intensity based on the listeners activity. I bet in real world tests this app would generate slow, wobbly stuff because it relies on the user activity to step up. If a user lowers activity, the music needs to ramp up, not down! It’s no coincidence that the BPM of most dance music is twice the human heart rate.

    Like or Dislike: Thumb up 0 Thumb down 0

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>