Financial statement fraud is a significant concern for investors, regulators, and financial institutions. Detecting such fraudulent activities is crucial for ensuring financial integrity and maintaining trust in the financial system. This research paper explores the application of natural language processing (NLP) and sentiment analysis to detect financial statement fraud by analyzing textual data within financial reports. Specifically, we focus on three companies with a history of financial statement fraud: Wirecard, Tesco, and Under Armor. By applying sentiment analysis to financial reports, we aim to identify linguistic cues that may signify fraudulent behavior. The results of our analysis show that NLP can be a valuable tool in detecting financial statement fraud, as it can uncover patterns and anomalies in financial reports that may indicate fraudulent activities. In this research paper, we use spaCy and spacytextblob NLP techniques to analyze the sentiment scores. We calculate the polarity and subjectivity scores for each report and compare them to previous research. Incorporating the Spacy + SpacyTextBlob method in our analysis yielded a significant improvement in detecting linguistic cues indicative of financial statement fraud, showcasing a percentage increase in accuracy over the TextBlob method. This enhancement allowed for a more nuanced and precise identification of potential fraudulent activities within the financial reports of the selected companies.