TY  - JOUR
AU  - Kadu, Lukesh Rameshpant 
AU  - Deshpande, Manoj 
AU  - Pawar, Vijaykumar 
PY  - 2026
TI  - Vision Tune: A Deep Learning Framework for Sentiment Driven Video, Image and Music Creation
JF  - Journal of Computer Science
VL  - 22
IS  - 4
DO  - 10.3844/jcssp.2026.1476.1483
UR  - https://thescipub.com/abstract/jcssp.2026.1476.1483
AB  - Artificial intelligence has enabled powerful generative models for text, images, video, and music, yet most tools still operate independently without a unified, multi-modal workflow. This article proposes an integrated AI framework, Vision Tune, that consolidates these isolated capabilities into a single, sentiment-aware platform for end-to-end media creation. The system leverages deep learning and multi-scope AI models to automatically generate written content, images, videos, and music for both creative and analytical applications, while emphasizing scalability, modular design, and user-centric interaction. By supporting cross-domain media synthesis and sentiment-driven customization, the framework targets real-world use cases in marketing, education, entertainment, and content production, where coordinated multi-modal outputs can enhance engagement and productivity. Beyond unification, the work highlights how the proposed architecture advances current AI media pipelines by reducing tool fragmentation, enabling cross-modal consistency, and providing a foundation for future extensions such as real- time generation, personalization, and human AI collaborative creation.