Bi-Lingual Audio Detection for Sentiment Analysis

This study proposes a hybrid deep learning framework in the field of sentiment analysis of bilingual (Hindi-English) audio recordings, acknowledging the problems caused by code-switching in mixed language conversation. This model based on Bi Directional Long short-term memory (BiLSTM) and long short-term memory (LSTM) networks captures temporal and contextual patterns for the classification of emotions, such as happiness, fear, sadness, and neutrality. Results demonstrate the capability of the model to capture emotional nuances in code-switched audio; these abilities may enable further applications in multilingual environments, including virtual assistants and automated customer support systems. The research establishes a fundamental platform for addressing code-switching problems in audio sentiment analysis.