Abuse is becoming more common on social media platforms, which has a negative influence on internet users. Homophobia/transphobia is the fear, hatred, discomfort, or suspicion of lesbian, gay, transgender, or bisexual people. Homophobic/transphobic speech, which may be characterized as hate speech directed towards LGBTQ+ individuals, has become a major concern in recent years. Online homophobia/transphobia is a serious societal issue that may make online platforms toxic and unwelcoming to LGBTQ+ persons while also seeking to eradicate equality, diversity, and inclusion. The main objective of our project is to identify the homophobic and transphobic content from social media comments. We took three languages Tamil, Malayalam, and English for these three languages we performed various machine learning(ML) models and deep learning(DL) techniques. In these models, random forest(RF) with TF-IDF, XGBoost, and LaBSE model performed well with good accuracy.