Frame Level Speech Recognition
Application of feedforward neural networks to recognize phoneme states in audio recordings from the Wall Street Journal dataset.
Posts by Collection
portfolio
MyTorch
MyTorch is a python Deep Learning framework. Implemented using Numpy, the framework supports functionality for building and training deep learning networks.
Multilingual Automatic Speech Recognition for Kinyarwanda, Swahili, and Luganda: Advancing ASR in Select East African Languages
We presents a multilingual ASR model for Kinyarwanda, Swahili, and Luganda, achieving a WER of 21.91 using a 3,900-hour curated dataset from the Common Voice project.
Download here
JSE Stock Trends Prediction
We model and predict Johannesburg Securities Exchange (JSE) stock trends using Markov Chains and benchmark them against popular machine learning techniques.
Download here
Comparison of Deep Reinforcement Learning Algorithms in Enhancing Energy Trading in Microgrids
We propose using microgrids powered by renewable energy and enhanced with machine learning for energy trading to improve electricity supply in rural Sudan.
Download here
Textual Inversion for Personalization of Diffusion-Based Image Generation.
We use textual inversion to adapt the Stable Diffusion model for text-to-image generation, fine-tuning it with a dataset including LAION 5B and images from African artists to generate culturally specific artistic styles, achieving above-average performance in evaluations.
Download here
publications
Arena-Bench: A Benchmarking Suite for Obstacle Avoidance Approaches in Highly Dynamic Environments
Published in Robotics and Automation Letters (RA-L), 2022
The ability to autonomously navigate safely, especially within dynamic environments, is paramount for mobile robotics. In recent years, DRL approaches have shown superior performance in dynamic obstacle avoidance. However, these learning-based approaches are often developed in specially designed simulation environments and are hard to test against conventional planning approaches. Furthermore, the integration and deployment of these approaches into real robotic platforms are not yet completely solved. In this paper, we present Arena-bench, a benchmark suite to train, test, and evaluate navigation planners on different robotic platforms within 3D environments. It provides tools to design and generate highly dynamic evaluation worlds, scenarios, and tasks for autonomous navigation and is fully integrated into the robot operating system. To demonstrate the functionalities of our suite, we trained a DRL agent on our platform and compared it against a variety of existing different model-based and learning-based navigation approaches on a variety of relevant metrics. Finally, we deployed the approaches towards real robots and demonstrated the reproducibility of the results.
Recommended citation: L. Kastner et. al. (2022) "Arena-Bench: A Benchmarking Suite for Obstacle Avoidance Approaches in Highly Dynamic Environments" Robotics and Automation Letters. https://arxiv.org/abs/2206.05728
talks
Talk 1 on Relevant Topic in Your Field
Published:
This is a description of your talk, which is a markdown files that can be all markdown-ified like any other post. Yay markdown!
Conference Proceeding talk 3 on Relevant Topic in Your Field
Published:
This is a description of your conference proceedings talk, note the different field in type. You can put anything in this field.
teaching
Teaching experience 1
Undergraduate course, University 1, Department, 2014
This is a description of a teaching experience. You can use markdown like any other post.
Teaching experience 2
Workshop, University 1, Department, 2015
This is a description of a teaching experience. You can use markdown like any other post.