This study examines the ability of CEO speeches to predict dividend declarations within the Jordanian industrial sector. The research employed two machine learning methods, logistic regression and neural networks, using 430 speeches from 33 publicly listed companies on the Amman Stock Exchange from 2018 to 2022. To achieve this purpose, several text-mining methods were utilized, including Term Frequency-Inverse Document Frequency (TF-IDF) and n-grams, to extract features as predictors of dividend outcomes. The results indicated that CEO speeches contain significant signals related to future dividend decisions. Moreover, the neural network model outperformed logistic regression, achieving 85% accuracy. The findings of this research demonstrate the significance of corporate communication, specifically CEO speeches, in shaping financial decisions in emerging markets. Additionally, it suggests that the integration of qualitative data with traditional financial metrics can help financial analysts and investors make more accurate financial forecasts.