https://github.com/absarraashid3/visionwhiz

VisionWhiz is a futuristic AI-powered web application that allows users to ask questions about any image using voice or text—and receive intelligent answers, both spoken and displayed. By integrating powerful models from OpenAI, Salesforce, and Google, VisionWhiz delivers real-time, context-aware visual interaction through a sleek Gradio interface.
https://github.com/absarraashid3/visionwhiz

Last synced: about 1 year ago
JSON representation

VisionWhiz is a futuristic AI-powered web application that allows users to ask questions about any image using voice or text—and receive intelligent answers, both spoken and displayed. By integrating powerful models from OpenAI, Salesforce, and Google, VisionWhiz delivers real-time, context-aware visual interaction through a sleek Gradio interface.

• Host: GitHub
• URL: https://github.com/absarraashid3/visionwhiz
• Owner: AbsarRaashid3
• Created: 2025-06-24T19:34:53.000Z (about 1 year ago)
• Default Branch: main
• Last Pushed: 2025-06-24T19:41:41.000Z (about 1 year ago)
• Last Synced: 2025-06-24T20:41:36.208Z (about 1 year ago)
• Language: Python
• Size: 5.86 KB
• Stars: 0
• Watchers: 0
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# VisionWhiz: Talk to Your Image AI

VisionWhiz is a futuristic AI-powered web application that allows users to ask questions about any image using voice or text—and receive intelligent answers, both spoken and displayed. By integrating powerful models from OpenAI, Salesforce, and Google, VisionWhiz delivers real-time, context-aware visual interaction through a sleek Gradio interface.

#  Project Overview

The goal of VisionWhiz is to:

Enable natural, multimodal interaction with images through voice and text.

Seamlessly integrate speech recognition, visual question answering, and text-to-speech features.

Improve accessibility and elevate the human-computer interaction experience.

#  Features

Voice & Text Input

Speech-to-Text: Speak your question naturally (powered by openai/whisper-small).

Text Input: Optionally, type your question into the input box.

#  Visual Intelligence

Image Upload: Upload any image for interactive questioning.

Visual Question Answering: Get accurate, context-aware answers using Salesforce/blip-vqa-base.

#  Smart Audio Output

Text-to-Speech: Listen to the answer read out loud with Google’s gTTS.

# Interactive Interface

Gradio Frontend: Built with Gradio and custom CSS for a clean, modern UI.

Smooth, real-time interaction across all components.

# Tech Stack

Component	Model/Library

Speech-to-Text	openai/whisper-small (HuggingFace)

Image QA	Salesforce/blip-vqa-base

Text-to-Speech	gTTS (Google Text-to-Speech)

UI Framework	Gradio

Audio/Image I/O	torchaudio, soundfile, PIL

# Project Structure

bash

Copy

Edit

VisionWhiz/

├── app.py               # Main app logic and Gradio interface

└── README.md            # Project documentation

#  Use Cases

 Ask what’s happening in any image using natural language.

 Create accessible educational tools with audio support.

 Aid users with visual impairments or reading difficulties.

 Prototype intelligent assistants for vision-language tasks.

# ARCHITECTURE DIAGRAM

![1_FXFCM5IrtGG3xyTr97-ONg](https://github.com/user-attachments/assets/df481120-acf2-41e5-83b9-f10ce6300292)

# Performance Snapshot





# Developed by

Absar Raashid