Speech recognition (SR) is a technique that uses a variety of identification and conversion algorithms to identify spoken words and translate them into text. It functions as a foundational machine learning application. There are several analyses and applications that can be performed with the converted content. The Google Speech Recognition API is used in this project because of its high level of performance. Another method that extracts and transforms data from a dataset to identify emotions from text is called sentiment analysis (SA). Several classification techniques, such as Support Vector Machines (SVMs), Naive Bayes, and Linear Regression, can be used in sentiment analysis. By comparing extracted words with datasets, the main goal is to recognize and categorize emotions such as joy, sadness, happiness, anger, frustration, and fear. The Naive Bayes model has been trained on a few sample speech inputs to eventually if it’s a question or a command. To further identify the text’s polarity and categorize it as positive, neutral, or negative, sentiment analysis is applied. These are established using the Text Blob library’s polarity score, which varies from −1 to 1. Negative: polarity <0 (e.g., −0.3), Positive: polarity >0 (e.g., 0.5), Neutral: polarity = 0. This project emphasizes the useful application of Python modules for audio processing, classification, and analysis. It shows how machine learning and natural language processing may be integrated for real-time voice recognition and sentiment analysis.