DweshaMukt leverages BERT and deep learning to detect hate speech in Hinglish, Hindi, and English across text, audio, video, images, GIFs, and YouTube comments. With real-time analysis, emoticon detection, and a Streamlit interface, it aims to foster safer online spaces through advanced NLP techniques.
multilingual gifs text images videos google-vision-api telegram-bots youtube-comments multimodal audios hate-speech-detection streamlit speech-to-text-api project-copyright project-funding google-video-intelligence-api ieee-research-paper
- Updated
Nov 27, 2025 - Jupyter Notebook