Inspiration
We felt that it might be a bit challenging for a new creator to break into the field with limited technical skills or could find it hard to find music of their liking for a video.
What it does
MoodMaker helps creators by generating music based on an uploaded video, helping to... set the mood! Furthermore it also helps to automatically add this music to the video background, saving the hassle of cutting or looping audio.
How we built it
We used VueJS as our frontend and Flask to handle our backend. We feed frames of the video provided to gpt4o for it to generate a prompt, the prompt is then sent to Suno AI which will use this prompt to generate a melody.
Challenges we ran into
Deployment, as retrieving our files currently takes up too much memory therefore this project can only be currently be run locally and the user will have to wait a considerable amount of time (ard 5mins). Suno AI that we are using to generate the music has limited generations per day (up to 2 songs generated).
Accomplishments that we're proud of
That we even submitted this on time.
What's next for MoodMaker
Giving users more customisation and freedom during the music generation. Getting the Heroku deployment to work.
**Please note there is a limited amount of tries as the gpt4o api is only running on a limited amount of credits.
Built With
- chatgpt
- flask
- javascript
- python
- suno
- vue
Log in or sign up for Devpost to join the conversation.