talk-data.com

The aim of the AI Music Mentor project is to help beginner electronic music producers finish their tracks by providing personalized, genre-specific, and actionable feedback that mimics the guidance of an experienced producer. This is achieved using LLMs with RAG, audio embeddings (CLAP/OpenL3), and Librosa analysis, focusing primarily on arrangement and structure while offering practical next-step suggestions.

Digital Twin Platform with Sensor Fusion Models integrates multimodal data streams (visual, thermal, RADAR, LIDAR, telemetry) into an interactive real-time dashboard for disaster management and defense. Built on the CARLA simulation framework, it enables detection, classification, and tracking of entities through sensor fusion, enhancing situational awareness with alerts, trajectories, and contextual insights. For the demo, the focus is on visual + RADAR fusion, with future extensions to live sensor feeds. The aim of the project is to create a digital twin platform powered by sensor fusion that provides actionable, interpretable, and real-time insights in complex environments, overcoming the limitations of single-sensor systems for improved decision-making.

The aim of the project is an educational platform that helps women and their partners understand mental and physical changes throughout the menstrual cycle. By tracking basal body temperature (manually or via API) and daily mood, the app predicts cycle phases and provides personalized, scientifically based recommendations for nutrition, supplements, activities, and mood management, accounting for interactions with medications or hormonal treatments. It consolidates multiple functionalities into a single platform, reducing the need for several apps while offering detailed explanations for each suggestion.

EEG-Based Motor Imagery Detection with Musical Feedback for Accessible Neurorehabilitation is an affordable brain–computer interface (BCI) that transforms EEG signals into spectrograms to classify imagined limb movements using CNN or CNN–LSTM models. Each detected motor imagery triggers mapped musical feedback, creating engaging sound textures that enhance motivation, neuroplasticity, and adherence while doubling as a creative tool. Potential extensions include reinforcement learning for adaptive, personalized musical feedback, balancing therapeutic benefit with user engagement.