Spandan Bandhu

Final-year B.Tech (CSE - Data Science) student with an 8.78 GPA. Skilled in ML, Deep Learning, NLP, Java, and Python. Experience as a Software Trainee at CyberSwift Infotech, with past roles in ML and Data Science. Worked on ML projects and presented research papers.

• A Hybrid Deep Learning Framework for Text Independent Automatic Speaker Recognition System, This research work focuses on the recognition of speakers automatically. The research achieves 16-17% Error Rate for 'LJspeech' dataset. This research paper was orally presented at ICACA'24 conference., https://link.springer.com/chapter/10.1007/978-981-97-4799-3_15
• An Efficient Deep Convolutional Neural Network for Automated Bird Song Classification (Under Publish Procedure), A custom CNN architecture has been implemented for bird song classification. The model achieves accuracy of 94.46% for 'Xeno-Canto' dataset and 97.02% for 'British Bird Song' dataset. This research work was presented verbally at ICDAI'24 conference.