Skip to content

Repository files navigation

ChattyStickers — AI Animated & Talking Stickers

License: MIT EARCP Verified

ChattyStickers is an open-source, AI-powered pipeline that transforms simple text phrases into high-quality, animated, and talking stickers. It uses state-of-the-art models for image generation (FLUX), text-to-speech (gTTS), and motion AI, all validated by the EARCP multi-expert algorithm for perfect coherence.

Pirate Cat Demo

🚀 Native Innovation: Talking Stickers

Unlike standard silent stickers, ChattyStickers generates unified WebM (VP9/Opus) files that embed both high-quality animation and synchronized AI voice.

  • Primary Format: WebM (Multimodal)
  • Fallbacks: Animated GIF (Silent), Animated WebP (Silent)
  • Quality: FLUX.1 (Schnell/Dev) for premium visuals

🛠️ Features

  • Multimodal FUSION: Generates image, voice, and animation in sync.
  • EARCP Validation: Real-time scoring of quality and coherence (Text vs. Image vs. Audio).
  • English/French Support: Automatically detects and translates your intent.
  • Web Interface: Clean, modern UI for instant generation and multi-format download.
  • Open Source: Built with Flask, PIL, OpenCV, and HuggingFace API.

🏗️ Architecture

graph TD
    A[User Phrase] --> B[Text Parser]
    B --> C[Image Generator - FLUX]
    B --> D[TTS Engine - gTTS]
    C --> E[Animator - Lip Sync & Motion]
    D --> E
    E --> F[Unified WebM Export]
    F --> G[EARCP Verifier]
    G --> H[Final Export: WebM, GIF, WebP]
Loading

📦 Installation & Setup

1. Clone the repository

git clone https://github.com/Volgat/chattystickers.git
cd chattystickers

2. Install dependencies

Ensure you have Python 3.10+ installed.

pip install -r requirements.txt

3. Configure Environment

Create a .env file in the root directory:

HUGGINGFACE_TOKEN=your_hf_token_here

4. Run the application

python app.py

Open http://localhost:5000 in your browser.


🖼️ Usage Instructions

  1. Describe your sticker: Enter a phrase like "A cool cartoon pirate cat laughing loudly on treasure".
  2. Generate: Click the "Generate Sticker" button.
  3. Wait for Pipeline:
    • 🧠 Analyzing text...
    • 🎨 Generating AI Image (SDXL/FLUX)...
    • 🔊 Voice Synthesis TTS...
    • 🎬 Animating sticker...
    • 🛡️ EARCP Verification...
  4. Preview & Download: Listen to your sticker and download it in your preferred format.

Process Screenshot


🛡️ Coherence Verification (EARCP)

Every sticker is validated by the EARCP (Expert Agreement for Reliable Coherence Parsing) algorithm. It measures:

  • Component Scores: Quality of individual media (Image, Audio, Animation).
  • Inter-component Coherence: Agreement between the text intent and the visual/auditory result.

EARCP Score Results


🤝 Integration

You can integrate ChattyStickers into your own discord/telegram bots via our API:

Endpoint: POST /api/generate Payload: {"phrase": "My phrase"} Response:

{
  "success": true,
  "sticker": {
    "url": "/output/abc/sticker.webm",
    "type": "webm",
    "has_audio": true
  },
  "earcp_report": { "score": 0.99, "grade": "A" }
}

📜 License

Published under the MIT License. See LICENSE for details.

Developed with ❤️ by Mike Amega powered by EARCP AI.

About

No description, website, or topics provided.

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages