Emotion detection and sentiment analysis are crucial parts of natural language processing (NLP). They help us understand the feelings expressed in written text. While there has been significant progress in analyzing sentiments in major languages, less-resourced languages like Hindi face difficulties due to a lack of linguistic tools and data. Our work aims to address the creation of corpus for Hindi language to enhance the accuracy of emotion detection and sentiment analysis. This involves collecting diverse text sources, cleaning, and organizing the data, and enriching it with manual and automated annotations for linguistic features and sentiments. Our work starts by collecting a set of Hindi words that are labeled as positive, negative, or objective. To make this collection richer, we add synonyms using IndoWordNet, which contains many word relationships in Hindi. We then translate these expanded sets of words from English to Hindi using advanced tools, ensuring the emotional meanings are accurately preserved. Each word in our dictionary is given a sentiment score using SentiWordNet, a tool that assigns sentiment scores for various languages. We normalize these scores to ensure consistency and accuracy across different word types and their uses. By focusing on creating a comprehensive corpus for Hindi, this study fills a crucial gap in resources and provides better insights into emotions in Hindi text. The sentiment dictionary we created serves as a solid foundation for further research and practical applications in NLP.