Courses and Thesis

This course focuses on the fundamental techniques used in the analysis and processing of audio signals. Students learn about the representation of sound in digital formats and the mathematical tools used to process and manipulate audio data. Topics covered include Fourier transforms, spectral analysis, and time-frequency analysis techniques for analyzing the frequency components of audio signals. The course also delves into filtering techniques, such as low-pass, high-pass, and band-pass filters, and their applications in audio enhancement and noise reduction. Advanced topics include feature extraction for speech and music analysis, such as MFCC (Mel Frequency Cepstral Coefficients) and chromagram, and their use in speech recognition and music classification. Students also explore source separation methods, like blind source separation (BSS), and dynamic range compression for improving audio quality. The course includes hands-on projects that apply these techniques to real-world tasks such as speech synthesis, audio classification, and audio segmentation, providing students with practical experience in processing and analyzing audio signals for a variety of applications.

This course focuses on applying machine learning techniques to audio signal processing. Students learn how to process, analyze, and classify audio data using various machine learning algorithms. Key topics include feature extraction methods like MFCC (Mel Frequency Cepstral Coefficients) and spectrograms, which convert raw audio signals into meaningful representations for further analysis. The course covers supervised learning techniques, such as classification and regression, applied to tasks like speech recognition, music genre classification, and speaker identification. Students also explore unsupervised learning methods, including clustering and autoencoders, for tasks like audio segmentation and anomaly detection in audio signals. Advanced topics include the use of Recurrent Neural Networks (RNNs), Long Short-Term Memory (LSTM) networks, and Convolutional Neural Networks (CNNs) for handling temporal and spatial patterns in audio data. The course also introduces generative models, like Generative Adversarial Networks (GANs), for audio synthesis and enhancement. Hands-on projects give students experience applying machine learning models to real-world audio tasks, such as speech-to-text conversion, emotion detection, and sound event detection, providing a comprehensive foundation in machine learning for audio signal processing.

This course explores the application of artificial intelligence (AI) in the healthcare industry, focusing on how AI can improve medical diagnostics, treatment planning, and patient care. Topics include machine learning techniques such as supervised and unsupervised learning for tasks like disease classification, image recognition in medical imaging (e.g., CT scans, MRI, X-rays), and predictive modeling for patient outcomes. Students learn about the development of AI models for personalized medicine, drug discovery, and patient monitoring through sensor data. The course also covers the ethical and regulatory challenges in AI healthcare applications, including data privacy, bias mitigation, and the integration of AI into clinical decision-making processes. Key machine learning algorithms, including deep learning methods like Convolutional Neural Networks (CNNs) for image analysis and Recurrent Neural Networks (RNNs) for time-series data from patient monitoring, are covered. Additionally, the course emphasizes clinical validation, model interpretability, and how AI can support doctors and healthcare professionals in making data-driven decisions. Hands-on projects allow students to work with real-world medical datasets, applying AI techniques to improve diagnostic accuracy and healthcare outcomes.