Sugran Maharastrian Recipe Generator

Read full text Summarise with VIRAT
Authors: Sujal Baviskar, Prachi Bhombe, Nikita Bharambe, Amit Khatal, Rajani Amol Hemade
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Maharashtra is known for its rich culinary heritage, representing a diverse range of fla- vors, ingredients, and cooking traditions across regions such as Pune, Kolhapur, Nashik, Vidarbha, and Konkan. Each region offers distinct variations of popular dishes, shaped by local ingredients, culture, and traditional cooking practices. Despite this diversity, there is limited digital documentation that accurately preserves or generates these regional recipe variations. With the rapid advancement of Artificial Intelligence, especially multimodal models capable of interpreting both images and text, new opportunities have emerged to digitize, preserve, and analyze regional cuisines intelligently. The project titled ”Sugran – Maharashtrian Recipe Generator” aims to create an AI- powered system that identifies a Maharashtrian dish from an uploaded image and gen- erates region-specific recipe variations. By combining computer vision and large lan- guage models, the system bridges the gap between cultural food heritage and modern AI-driven technology, ensuring that authentic recipes and culinary knowledge are pre- served digitally. Most recipe applications available today offer generic, standardized versions of dishes. They often fail to incorporate cultural variations and regional nuances that are essential to Indian cuisine, particularly Maharashtrian dishes. For example, a single dish like ”Misal” has multiple well-known variants including Puneri Misal, Kolhapuri Misal, and Nashik Misal—each distinguished by spice levels, ingredients, texture, and preparation techniques.

Fields of Study & Subjects

Culinary Culture and Tourism Advanced Neural Network Applications Computer Science and Engineering

Publication details

Journal / source
Zenodo (CERN European Organization for Nuclear Research)
Publication date
21 March 2026
Language
ANG