@andknf already broke it. Do you remember moodlogic? Similar concept of trying to map songs via "mood." They had major data integrity issues. Are you using Echonest or something to determine what kinda mood a song is?
@micah I'm sorry to hear that man! Can you send me a description of what happened when it exploded? I would be very grateful, as I'm a one-man show and need all the eyes on tech issues I can get. andrew@strings.fmnnI wish I could say I used Moodlogic, I would love to know how they did things and what they learned. Did you use it enough to have any favorite features? And yep, currently, we do link songs up using Echonest.
@andknf no - was commenting that you broke producthunt's emoji support. I used ML years and years ago. The biggest issue was song profiling. There is a Spotify app that does a decent job, but honestly the best "mood" playlists have always been hand curated. Curious as to the tech.
@micah hah phew! Bug squad on lower alert. Aura, the algorithm that powers the app, uses lots of different song features as well as user event data to rank moods and then determine if they've been ranked well. nnBeyond the tech, what makes a string meaningful is that you do curate enough of it that it says something about who you are. We just help you so that you don't have to pick 20 tunes to create a playlist you love, leaving you free to enjoy the moment you're in.
Hello Product Hunters! I'm here to read your mood and answer questions about strings.fm and our wonderful mood-learning algorithm, Aura. I'm also going to test the limits of Product Hunt's emoji-display capabilities! 💙😽😎
Blurb
No More
Blurb
No More
No More
Product Hunt
No More
The Fundraising Playbook