CrowdEmotion Case Study: Spotify Music Mashup Video

Ever imagined about having an automated music playlist that reads your current emotional status? Well now, this future is getting real.

In partnership with Spotify, CrowdEmotion deployed facial coding to develop the world’s first emotion-driven music discovery platform. Spotify currently uses a database analysis system to uncover their users’ music preferences. Hence, Crowd Emotion, exclusively for Huddle, explored the possibility of an emotion-driven music recommendation system. To do so, we focused on what people’s emotional engagement indicate about their music preferences.

Method

Listeners first did a questionnaire about their musical preferences. They watched a mashup of indie artists to measure their emotional engagement while exploring new music. Participants received real-time report results of their emotions during the video and a full report of their overall emotional engagement.

Each participant was given a persona profile comparing their emotional engagement towards different music genres with their preferred music genres. Individual results displayed that people often engaged with music genres beyond their preferred genres. We made 97 emotional discoveries in total and 87 potential new listeners, producing a lead conversion of 48%. That is one every 4:48 seconds!

Highlights

  • 97 Emotional Discoveries
  • 87 Potential new listeners
  • 48% Lead conversion
  • One lead conversion every 4:48 seconds!

Conclusion

All in all, this platform allowed Spotify to understand the musical preferences of users beyond what they claim to be preferred. Hence, this partnership allowed Spotify to uncover their listeners’ implicit musical taste, working towards shaping the world’s very first emotion-based music recommendation software.

One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.