title	emoji	colorFrom	colorTo	sdk	sdk_version	app_file	pinned	license	about	short_description
QWQ EDGE	💬	red	gray	gradio	5.15.0	app.py	true	creativeml-openrail-m	demo space to try how multi model selection function works	Multimodality

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

QwQ Edge 💬

QwQ Edge is an advanced multimodal chatbot that integrates AI-driven text-to-speech (TTS) generation, image generation using Stable Diffusion XL (SDXL), and a conversational interface powered by large language models. The app supports real-time conversations with multimedia output, including images and speech, and offers a flexible environment for creative and interactive communication.

Key Features:

1. Multimodal Conversational AI: Supports text, image, and speech as input/output.
2. Text-to-Speech (TTS): Convert text responses into speech with multiple voice options using Edge TTS.
3. Image Generation: Generate high-quality images based on prompts using the Stable Diffusion XL pipeline.
4. Multimodal Inputs: Handle text, image files, and spoken queries for generating appropriate outputs.
5. Real-time Interaction: Provides instant responses with streaming output and live updates for image generation and TTS processing.
6. Customizable Parameters: Fine-tune token generation parameters such as temperature, top-p, top-k, and repetition penalties for more controlled responses.

Technologies Used:

• Gradio: For creating the user-friendly interface.
• Transformers: For NLP and text processing (Hugging Face models).
• Stable Diffusion XL: For high-quality image generation.
• Edge TTS: For converting text into speech.
• Python: For the backend logic and model integration.
• PyTorch: For deep learning model inference and acceleration.

Supported Commands:

• Text-to-Speech (TTS):
  • @tts1: Use the "en-US-JennyNeural" voice.
  • @tts2: Use the "en-US-GuyNeural" voice.
• Image Generation:
  • @image <description>: Generate an image from the given description using Stable Diffusion XL.

Environment Setup

1. Clone the repository:

   git clone https://github.com/PRITHIVSAKTHIUR/QwQ-Edge.git
   cd QwQ-Edge

2. Install the required dependencies:

   pip install -r requirements.txt

3. Set up the environment variables for model paths and parameters:

   • MODEL_VAL_PATH: Path to the SDXL model.
   • MAX_INPUT_TOKEN_LENGTH: Maximum token length for model inputs.
   • Other environment variables for batch size, resolution, and CPU/GPU usage.

4. Launch the app:

   python app.py

Example Usage:

1. TTS Example:

   • Type: @tts1 What is quantum computing?
   • The app will convert this text into speech using the JennyNeural voice.

2. Image Generation Example:

   • Type: @image A futuristic city skyline at sunset with neon lights
   • The app will generate an image based on the prompt using the SDXL model.

3. Chatbot Conversation Example:

   • Type: What is the capital of France?
   • The app will respond with a chatbot-generated text response.

Technical Architecture

graph TD
    A[User Interface] --> B[Chat Logic]
    B --> C{Command Type}
    C -->|Text| D[FastThink-0.5B]
    C -->|Image| E[Qwen2-VL-OCR-2B]
    C -->|@image| F[Stable Diffusion XL]
    C -->|@tts| G[Edge TTS]
    D --> H[Response]
    E --> H
    F --> H
    G --> H

Loading