This is a project for the Microsoft learn student ambassadors December to February AI project. The project will be on building an app that uses Microsoft Azure vision, to get the descriptions of our images, with the following features.
- Generate image descriptions and allow you to ask follow-up questions.
- Allows you to communicate using speech, and here an audio feedback if you want to.
- Generate brief captions for images, suitable for the web.
- Take a screenshot to describe it.
- Describe any window, app, or whatever can be snapped. Including yourself using your device camera!
Just like Microsoft has seeingAI for Android and IOS, consider this as a seeingAI for windows.
- Firstly, you will need access to the Azure portal.
- You can follow this article to create a free account, or even create an azure for student account if you are a student.
- Then, create a speech and vision resource. For an instruction on how to create an azure resource, click here.. You should be fine after that.
- Get your keys and endpoint, then modify auth module in this project to suit your needs.
- Clone this repo to your local machine:
https://github.com/kefaslungu/neocasa.git - And then I'll suggest you create a virtual environment to store all the dependences for this project.
Pip install -r requirements.txt.
You don't need me to tell you that you need python install right? Well at list version 3.8 is required, but version 3.10 is recommended!
If you don't want to go through the stress of installing python and the stress of creating a virtual environment or even installing dependences, don't worry. I'll release binaries just for you!
But, you still need to get your keys and endpoint. Just follow the instruction of the app to insert those things if you are using the binaries.