Presenting Audio Imaginator

We were searching for ideas, reviewing different options and possibilities of the presented tools. We were considering two options: either something with audio transcription, or video generation using stable diffusion. We decided to merge those two ideas into one, and generate videos based on transcription from an audio.


Our application is a tool that allows users to upload an audio track and automatically generates a unique visual experience that is intended to enhance the listening experience and engage the audience. The application uses powerful algorithms to analyze the audio track and create a corresponding visual representation in real-time. The visual experience generated by the application could take many forms such as an animated music video, a visualizer that responds to the music, or a graphical representation of the audio track's key features


We've faced many difficulties during our process of development. The biggest one was the resources required to successfully generate a video. Generating the video is very time consuming, and in a competition such as this one, this doesn't leave much space for any mistakes. Another challenge for us was getting familiar with many new tools. Our team spent almost half of the time learning new things, and technologies. This is of course a great opportunity for us. Unfortunately this resulted in a poor graphical interface. We don't have much experience in front-end development.


We are really proud of that we have a chance to create fully functional app in such short time.


Summary

During this hackathon we learned how we can fuse different engines into one big application. Also we had opportunity to check many different tools during a development phase. Audio Imaginator is an open source project that we will want to use in our portfolio in the future. We planned to add more functionality and possible integrity with many video/social media platforms like YouTube, Instagram, Facebook or TikTok.


Built With

  • assembly-ai
  • python
  • stable-diffusion
Share this project:

Updates