The first of everything is always a learning experience - like dipping your toes into the water to get a feel of the tempature before doing a complete cannonball into the water on a hot sunny day. So for us, the first week was to dip our toes and meet the team and understand the goals of the project as well as what the next 10 weeks will look like since we were going to be working remotely from our homes.

Once we understood the gameplan, me and my co-intern, Mike Tota, started to set up what we needed to since we all were working virtually - like connecting to Gallaudet University’s VPN and the servers through ssh terminal and SAMBA to access the directories we use for our project. Then we got started on the literature review and did research on commerical products and their captioning set up, WebRTC and how connections happen, and all of the commerical Speech-To-Text APIs. That didn’t take us long so we moved on to finding 4 videos and audio to use for our controlled experiements and installing 3 Automatic Speech Recognition software -Google GCP, MS Azure, and IBM Watson- on our server with the help of our mentor, Dr. Raja Kushalnagar.

After finding 4 videos - we started to set up an enviroment in WebRTC to show the videos through live streaming with “Video to Peer Connection” method to be able to test our ASRs. At first, it was a struggle with the coding of JavaScript but with Norman Williams’ help - we were able to set up the enviroment. After that, me and Mike worked on getting familiar wit the ASRs in our server and tested them with audio samples. We ended the week with a team meeting, discussing our progress as well as the questions me and Mike had after our first week with the project, and a touch of what we will be doing next week.