A Bangla Speech Dataset for Sentence Type and Emotion Classification Using Traditional Machine Learning

Mahmudul Haque Shakir and Md. Saef Ullah Miah designed the experimental pipeline, prepared and trained the machine learning models, and carried out the evaluation, results analysis, and discussion. Aidah Anah, Tirtha Das Tanu, Md Imamul Islam, and Ahmed Al Mansur contributed to the construction of the Bangla speech dataset, including recording collection, annotation, and preprocessing, and to the preparation of the manuscript. All authors reviewed and approved the final version.