AI-Dermatologist is an AI-powered medical assistant that combines vision, speech-to-text, and text-to-speech capabilities to simulate a professional doctor–patient interaction. It enables users to record their voice, submit medical images, receive diagnostic insights, and hear spoken responses.

• Voice Interaction: Record patient audio via microphone and transcribe using Groq's Whisper model.
• Image Analysis: Process and evaluate medical images (e.g., skin lesions) using a multimodal LLM (Meta LLaMA Scout).
• AI-Generated Responses: Generate concise, doctor-style text responses without AI disclaimers or formatting.
• Text-to-Speech: Convert the AI-generated diagnosis into natural-sounding speech via gTTS (Google) or ElevenLabs.
• Web UI: Intuitive Gradio interface for seamless voice and image input, and playback of responses.

1. Audio Capture: User records their voice; saved as MP3 via speech_recognition and pydub.
2. Transcription: Transcribe audio to text using Groq Whisper (whisper-large-v3).
3. Prompt Assembly: Combine system prompt with transcription for context.
4. Image Encoding & Analysis: Base64-encode user image and query Meta LLaMA Scout via the Groq API.
5. Response Generation: Receive doctor-style advice from the LLM.
6. Speech Synthesis: Generate and play back spoken response using gTTS or ElevenLabs.
7. Web Interface: Gradio serves as the frontend for input/output.

• Python 3.8+
• Gradio: Web UI framework for rapid prototyping of ML interfaces.
• Groq SDK: For image and audio transcription API calls.
• gTTS & ElevenLabs: Text-to-speech engines.
• SpeechRecognition & pydub: Audio recording and format conversion.
• dotenv: Environment variable management.
• winsound / afplay / aplay: Cross-platform inline audio playback.

1. Clone the repository:

   git clone https://github.com/pranav-here/AI-Dermatologist.git
   cd AI-Dermatologist

2. Install dependencies:

   pip install -r requirements.txt

3. Create a .env file with the following keys:

   ELEVENLABS_API_KEY=<your_elevenlabs_api_key>
   GROQ_API_KEY=<your_groq_api_key>

Launch the Gradio interface:

python gradio_app.py

Open the URL printed in your console (typically http://127.0.0.1:7860).

• Step 1: Record your voice and/or upload an image.
• Step 2: View the transcribed text and AI-generated diagnosis.
• Step 3: Listen to the spoken response.

All dependencies are listed in requirements.txt. Example:

grpcio
gradio
pydub
speechrecognition
gtts
elevenlabs
groq-sdk
dotenv

Refer to requirements.txt for the full list and exact versions.

This project is licensed under the MIT License. See LICENSE for details.