chatbot_project_NTI

Bl_Araby(بالعرربي)

This Project is consisting of 3 phases, the first is Arabic Text and Audio summarize.

Secondly Arabic image caption, That describe what is in image.

Model Architecture

MobileNetV2

MobileNetV2 is a lightweight deep learning model that is highly efficient and suitable for mobile and edge devices. It serves as the feature extractor in our architecture.

Encoder-Decoder

The encoder-decoder model consists of:

• Encoder: Takes the feature vectors produced by MobileNetV2 and processes them for further decoding.
• Decoder: Converts the encoded features into human-readable descriptions.

Finally Chabot that It allows us to talk to the text, image, or audio file, and we can ask it

anything we want about them