Video Learner Insights (Use Case 2)

Team Split

1. LOGIN: Create a login component that allows a user (parent, therapist, etc...) to login into the application. This can be through social media, or through AAD. Incorporate security best practices. Store all user data into a database.

2. UPLOAD: Be able to upload a video file into an application. Store multiple videos along with their metadata into a database. View all the videos on the front end by session date. Optional: organize by month, year.

3. INTERACTION: Add functionality to comment at different timestamps of a given video. (Ex: 0:23 - Penny smiled when the toy was introduced.) Allow users to comment in real time. For each video, have a place to write down the session goals. Store all of this data in a database.

4. INSIGHTS: Allow users to view a transcript from the session video using Speech to Text. Extract insights of key words and sentiment on the video using Language services from Cognitive Services. In addition, create a summary of each session.

5. DOCUMENT: Create an architecture diagram of the Azure technologies we will use and document how each step connects with one another in our Github page.

Technology

• Azure AD B2C
• Azure Cogntiive Services
• Sharepoint on Stream
• Azure SQL
• Azure Functions

Pre-reqs

1. VS Code

Resources

• LOGIN

1. Sign in using AAD B2C

• INTERACTION

3. Overview of Stream on Sharepoint
4. Video Portal with Stream on Sharepoint
5. Video Portal with JSON formatted library

• INSIGHTS

1. Video Indexer API
2. Video Indexer Output
3. Azure Speech SDK
4. Speech To Text from a file
5. Key word extraction
6. Sentiment Analysis
7. Summarization