Ability of the deep learning methodologies to automatically extract complex features from waveforms, enable precise segmentation and analysis in diverse domains, and ultimately contribute to advancements in fields like healthcare, communication, and audio processing. Transfer learning and data augmentation strategies enhance model performance, addressing challenges posed by limited labeled datasets. Unsupervised learning approaches, including clustering and autoencoders, contribute to the discovery of underlying patterns in waveform data. Hybrid models, combining deep learning architectures with traditional signal processing techniques, offer a holistic solution, balancing interpretability, and domain knowledge. The study underscores the significance of these methodologies, emphasizing their role in advancing fields such as healthcare, communication, and audio processing through precise waveform segmentation. Overall, this comprehensive overview provides insights into the state-of-the-art techniques driving the evolution of deep learning-based waveform segmentation.