Inspiration
My inspiration for this project came from a multitude of sources. To begin with, I have been accompanied by music for almost my whole life. As I matured and gained more experience in the real world, I truly saw the importance of music to one's physical and mental health. Music has especially contributed to my mental well-being throughout the years, and I am very grateful for that. Hence, I have been increasingly interested in finding ways to share the joy music can bring to those around me. Not only does music improve one's mental health, it also increases one's productivity in studying. Furthermore, my main inspiration for this project, ironically, actually came as a byproduct of my brainstorming for this competition. What I mean is that whenever I sit down and get ready to study (in this case to brainstorm), I always search the internet, whether it be youtube or spotify, for instrumental music to listen to while I study. The problem lies in the fact that often times, I take 30 minutes to an hour to find a playlist of songs that I like and fit my mood for that study session. As a classical musician myself, I feel that instrumental music is often overlooked. While modern pop music, for instance, is categorized into "upbeat" or "experimental" or "indie", instrumental music is really only categorized to itself. However, this is far from reality. Instrumental music expresses emotions just as well, if not better than modern music. Thus, while it is extremely easy to find a playlist of "happy" pop music on the internet, I am puzzled as to why it is much more difficult to find a playlist of instrumental music that centers around a specific mood or theme. This became the main inspiration for me; I wanted a way to quickly find instrumental songs that matched my mood.
What it does
When you open the website, there will be a singular text box that asks for the user's study mood. The user will type his/her current mood and then click the music button right below. Study Mood will search for instrumental music that closely matches the input mood. Study Mood will then recommend 8 songs at a time that it determines to be a good match for the user's mood. It is important to note that Study Mood only considers instrumental music as the whole purpose of it is to find music to accompany the user while he/her studies; I purposefully programmed Study Mood so that it ignores any lyrical songs as many scientific studies have shown that lyrical songs reduces efficiency in studying.
How we built it
I built Study Mood using HTML, CSS, and Javascript. I have some experience with data analysis in Python. However, I have never taken a CS course, participated in a Hackathon, and used HTML, CSS, and Javascript (this is my very first time). The fundamental principle that Study Mood operates on is that it searches for instrumental music that matches a certain mood, which the user inputs. To do so, I found a csv on kaggle that included several thousands of instrumental songs all separated into a dozen unique acoustic features such as "danceability", "energy", or "loudness" etc. I imported the csv onto jupyter notebook and used Python to do some basic data cleaning. Then I set different parameters for each acoustic variable. The parameters were determined based on some research that categorized certain moods to given intervals of values for each acoustic feature. For instance, "happy" instrumental songs would imply that the songs would all have a danceability and energy factor of greater than 0.3/1.0 or 30%. Due to time limitations, I was only able to repeat this process for about 25-30 different moods/themes. Thus, as of status quo, Study Mood is only able to recognize about 25-30 moods.
Challenges we ran into
The biggest challenge that I ran into was just using HTML, CSS, and Javascript. My coding experience is very minimal. I've only used Python to do some basic data exploration before. HTML, CSS, and Javascript was unlike anything I've done before. Thus, I had to troubleshoot a lot and I mainly learned from watching YouTube videos and reading articles. Other than that, a more technical challenge was to determine the parameters for each acoustic variable that correlated with certain moods. This, in turn, made it difficult to program Study Mood to recognize more moods. Another thing is I tried to add an AI chatbot to my website. However, this was just too above my head, and I could not troubleshoot it correctly even after watching youtube videos and reading articles.
Accomplishments that we're proud of
I am extremely proud of successfully creating a website using HTML, CSS, and Javascript as before this weekend I did not even know what HTML and CSS were. I am also proud at the fact that I created something that is personally meaningful to me. Despite my minimal level in coding, I am very happy to persevere and use my creativity to create something that solves a real problem.
What we learned
I learned how to create a website in HTML, CSS, and Javascript. I also gained some insights into how "moods" can be quantified numerically in songs through different acoustic features.
What's next for Study Mood
In the very near future, I hope to train Study Mood to recognize more moods/themes. I hope to also incorporate new features to make the interface more interactive. For instance, I hope to add a timer for minutes focused and an AI chatbot that's able to provide support or advice on mental health for users. Currently, I have a timestamp and an inspirational quote that changes every time the user refreshes the page. This is because I think that many users including myself like to just stare into the computer screen: reading a quote might just provide some spark of motivation to keep on working. The whole purpose of Study Mood is not just to provide a way for users to quickly find a playlist of instrumental music that matches their mood for their study session, but I want to develop Study Mood into an extension like Momentum so that users will be able to essentially use Study Mood's webpage as an aesthetic home screen when they log onto the internet browser of their choice. Far down the line, I could also see Study Mood being made into an app. Similar to Spotify and Apple Music except Study Mood will be curated to only instrumental music; hopefully, it could become the primary app people go to to find instrumental songs.
Log in or sign up for Devpost to join the conversation.