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.
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
- 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.
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]
git clone https://github.com/Volgat/chattystickers.git
cd chattystickersEnsure you have Python 3.10+ installed.
pip install -r requirements.txtCreate a .env file in the root directory:
HUGGINGFACE_TOKEN=your_hf_token_herepython app.pyOpen http://localhost:5000 in your browser.
- Describe your sticker: Enter a phrase like "A cool cartoon pirate cat laughing loudly on treasure".
- Generate: Click the "Generate Sticker" button.
- Wait for Pipeline:
- 🧠 Analyzing text...
- 🎨 Generating AI Image (SDXL/FLUX)...
- 🔊 Voice Synthesis TTS...
- 🎬 Animating sticker...
- 🛡️ EARCP Verification...
- Preview & Download: Listen to your sticker and download it in your preferred format.
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.
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" }
}Published under the MIT License. See LICENSE for details.
Developed with ❤️ by Mike Amega powered by EARCP AI.


