Inspiration
No one wants to be handed the aux on a 5-hour road trip. What song do you pick? What artist? Immeasurable amounts of stress can arise from this one simple task. No one wants to be the one to kill the mood. CarJam seeks to resolve this universal issue for all of your future road trips and save some friendships. We strive to enable all listeners by truly embracing the idea of "going with the flow".
What it does
CarJam generates a unique and personal song queue by leveraging powerful speech recognition technology. The user is prompted to record ambient audio during their drive, be it a sentence or a whole conversation. This audio snippet is then processed via Google Speech to Text. The text is analyzed for general mood and split into factors such as danceability, tempo, etc. to parse the Spotify API for the perfect combination of songs. The user will then promptly receive a newly generated queue of songs to be played and can continue on their drive.
How we built it
We built the frontend with React and Material UI. It allows the user to record their speech and then plays the song. The audio recording is then sent to the backend, which was built using Node.js and Python. Using Spotify API, Google Speech-to-Text API, and Google Natural Language API, the last-cached sentiment data picks a song based on danceability, tempo, loudness, etc., and sends this song URL to the frontend.
Challenges we ran into
Some of the challenges we faced were: working around the Spotify API web token repeatedly expiring, learning object principles of javascript, and figuring out a way to route the audio recorded on the front end to be processed by the backend.
Accomplishments that we're proud of
We are proud of the fact that we were able to create an efficient and effective product. We met all of our goals and are particularly proud of the fact that we were able to successfully fetch audio and process it accordingly.
What we learned
Through making this project, we were exposed to JavaScript and using Spotify API.
What's next for CarJam
Some features we'd love to explore in regards to the future of CarJam include: • Implementing a Cloud API storage to cache previously loaded songs • Leveraging a machine learning-based algorithm to enhance text processing and song selection • Integrate an autoplay system to allow the users a hands-free option
Log in or sign up for Devpost to join the conversation.