Photos obtained, want to generate a playlist?
If your life was a movie, which would be the soundtrack of it? Tune my life is a platform which collects your Instagram photos and, after running computer vision on the photos, generates the perfect Spotify playlist that matches with your photos' story.
We wanted to develop something using sentiment analysis.
What it does
Tune my life collects your Instagram photos and, after running computer vision on the photos, generates the perfect Spotify playlist that matches with your photos' story.
How we built it
We built the platform with Python and used Flask for the server. The photos are obtained using the Instagram API and then processed with Microsoft cognitive services in order to apply the sentiment analysis and retrieve the data. Finally, the playlist is generated using the Spotify API according to parameters such as acousticness, danceability, loudness, tempo, balance, etc.
Challenges we ran into
API's restrictions (Pending approval of instagram to give public access), how to determine the music attributes from the parameters obtained when running cognitive analysis over the images, deploying to the server
Accomplishments that we're proud of
tl;dr -> It works! We have used some technologies that we hadn't use before and the results are pretty accurate.
What we learned
How to use Spotify's API, Instagram's and Microsoft cognitive services. Also we did some research related to the topic we discovered relations between feelings and music attributes.
What's next for Tune my life
- Get instagram's approval and be able to make it public.
- Use the images captions to obtain more information about the emotions related to the picture.