Nowadays, most peoples are using Internet and giving their opinion or sentiment on various issues, products, services, etc. Most of the time, the opinion or sentiment are expressed in written text along with the emoticons. There are various ways by which people are expressing their opinion or sentiment which is either having positive, negative, or neutral polarity. Due to availability of Internet and smartphone, huge data are generated which is needs to be analyzed, to make proper decision. The sentiment analysis process encompasses several critical stages: Gathering or retrieving data, pre-processing it, representing it as vectors with feature selection, and classifying sentiment using a range of algorithms such as machine learning (ML), deep learning (DL), lexicon-based methods, or hybrid techniques. This paper aims to review the latest research in sentiment analysis for Indian languages and to identify potential future directions to enhance existing methods or develop new approaches to address the various challenges and limitations in sentiment analysis for Indian languages.