Is to bridge the educational disparities that arises in traditional classroom paradigms.
Students classified to
- Slow, Mid, Fast learners on the basis of their IQ and understand on the subject
-
Demo App (v0.1)
- A lecture video (7m-1hr)
- AI is fed the lecture transcript
- Students are allowed to listen to the video
- Students can ask their doubts to the AI, and AI will answer in the ctxt of the video
-
Pre-MVP v0.2
- Additionally, Students would be shown the lecture snip that explains the answer to their q.
Outcome:
- We are not trying to replace human teachers
- We are trying to bridge the gap arises with students and human teachers.
-
MVP v0.3, v0.4
- Expand our demo video -> 5 videos list
- min: 90% accurate responses from AI in terms of lecture snip and prompt
-
MVP v0.5
- Add support for all youtube videos
-
MVP v1.0
- Role out RBAC (Role based Access control) to our beta users
- Roles of users: i) Teacher- Upload video, and transcript. Evaluate ii) Learner- Search bar: to search for a lecture video and watch it and ask qs
-
MVP v1.2
- Teacher Dashboard (see the video list, viewers etc) Implementation: look for open source dashboards - PowerBI
-
MVP v1.3
- AI Avatars powered by https://elevenlabs.io or https://studio.inworld.ai/ ......
- Explore various youtube videos
- Separate assistants for every video
- Backup assistant ids in redis with respect to video