New Progress on Final Product!

After talking to my mentor last week, we were able to go over some research papers, GitHub repositories, and other project information in order to understand the various approaches plausible to pursue my ISM Final Product. We found various datasets and algorithms that we could use, but we still decided to survey a few more research papers and projects online that demonstrate the applications of audio-visual networks for optimized speech recognition. I saw a few youtube videos and a brand-new approach of using transformers instead of traditional 3D CNNs or Transfer learning models. The past research has mainly served as a new understanding of the issue and has demonstrated that we can pursue many paths to make the Final Product successful.

This week, I hope to meet with my mentor Dr. Paschall and decide on the approach as well as the dataset we should use to start implementation. By completing this initial step, we will be well underway in the process of understanding past research and using it to optimize our solution with a new perspective.

You will have to wait for the next couple of weeks to see a decent prototype of the model and the live project. This extensive building process is definitely exciting and going well, so the wait will be worth it.

So see you next week, same place, same time.