Embodied Conversational Agents (ECA) seek to provide a more natural means of interaction for a user through verbal and non-verbal properties of human face-to-face communication. For this reason, these systems are found to bring benefits in different mental health related interventions. However, a key challenge in developing agents to replace the human interlocutor in a dyadic conversation, is to simulate appropriate attentive listening behaviors. In this thesis work, we explored different backchannel strategies and studied their effects in terms of likability and engagement. We built a fully embodied conversational agent with three different levels of backchannel strategies and ran a within-subject study with a convenience sample of 24 participants. The results showed that the amount of emotional words in the speech of users increased if the attentive listening capabilities of the agent were improved. In addition, the capability to trigger both verbal and nonverbal backchannels with proper timing was found to be a relevant feature in terms of improved speech rate and emotional words. Contrary to our hypothesis, backchannels based on actual emotion and sentiment analysis of the speech content were not found to be significantly influential on the quality of interaction. Multi-modal approaches are suggested for future works in order to overcome limitations of this work due to potential lack in emotion detection accuracy. ... Read more

In this thesis a cognitive multi-agent system is developed for the real-time strategy game StarCraft: Brood War using GOAL, a language developed at the Delft University of Technology. StarCraft provides various challenges to GOAL that it has not faced before in terms of complexity and quantity of agents. A connector is required for GOAL to interact with StarCraft: the StarCraft-GOAL Environment Connector. In order to provide context for the development stage, this thesis starts by explaining the basics of StarCraft, GOAL and the connector. The development of the bot that will be created, named ForceBot, takes place over the course of four tournaments that serve as milestones and are used to measure the game performance of ForceBot. As the conditions and participants of these tournaments vary, an additional setup for hosting internal test tournaments is established in order to provide a form of testing that remains consistent over the course of the development. Based on the results of tournaments, it can be concluded that the final game performance of ForceBot is above average with regards to existing StarCraft bots. A total of four aspects of GOAL, the language, tools, connector and computational performance, are analysed based on their usage and performance during development, and suggestions for improvements are made accordingly. Among these four aspects, the computational performance is identified as the primary bottleneck for the game performance of GOAL-based StarCraft bots. ... Read more

Games such as Skyrim use a narrative world, which is a game world that is used to tell stories. Creating narrative worlds is a time-consuming process because of the collaboration of writers, 3D artists and game designers, the story adjusting the world over time and the requirement for diverse and interesting environments. This thesis focuses on solving these problems. First, the manual approach was designed that allows the manual creation of a narrative world,
while focusing on filling environments. Secondly, the assisted approach builds on the manual approach and is able to assist a designer in making the decisions necessary to fill environments. The manual approach uses a set of objects with explicit relationships, created following a modular approach. The manual approach incorporates a discrete and sequential timeline of events, with each event connected to a single location, directly into the process. Several actions are used to fill the narrative world for each point in time of the timeline. The assisted approach offers assistance for each decision of the actions. Due to computational complexity, each decision is assisted independently. The discrete decisions are evaluated and ordered through the use of weighted criteria. The continuous decisions, the placement of objects, requires generation of options, which is done through an adaptation of Merrell et al. [2011]’s GPGPU algorithm. A usability study was conducted to test the effectiveness of the assisted approach compared to the manual approach. Results were collected through a creativity support index (CSI), a questionnaire on the assisted approach and quantitative measurements on object usage and time taken for actions and locations. The manual approach scored better in the CSI results than the assisted approach. The final questionnaire showed that this was caused by automation results not matching the user’s expectations as these differ, at least, per user and location type. The following conclusions are made. First, explicit
relationships allow direct insight in how objects interact. Secondly, the usability study results supports the iterative process employed by, both our, and many other creative process approaches. Finally, in general, it is difficult to provide valuable assistance as expectations differ greatly. The incorporation of uncertainty into the assistance can help lessen this problem. ... Read more

Sound spatialization is a natural, intuitive but sparsely researched topic in multi-track audio mixing. Although a lot of research has been devoted to the automatic fader gain settings, addition of dynamic range equalization and related effects, delay and Reverb have taken a backseat. The dichotomy in the artistic and engineering approaches to audio mixing have resulted in studio best practices not given their due with suitable algorithmic interpretations.
Due regard to studio practices along with a more holistic approach combining all the steps of audio mixing are especially necessary in the background of the exponential growth of bedroom studio producers and musicians, mixing and crafting their tracks personally. The additional growth in the availability of faster personal computing only fuels this trend.
This thesis attempts to be an exploratory foray into the addition of Reverb to production oriented multi-track mixing. Taking into account studio practices, 2 different algorithms are compared with a professionally mixed track and an unreverberated reference track. The results from hidden reference listening tests are analyzed to draw conclusions of the effectiveness of automatic methods of Reverb addition against the professionally mixed track.
The results suggest that the current algorithms implemented are unable to reach the subjective perceptual quality of the professionally mixed track. However, some important conclusions are drawn from the theoretical and experimental research which provide clear guidelines for possible future implementations. ... Read more

This thesis project worked towards extending the interactive capabilities of the RoboTutor Nao, by enabling it to answers natural language questions about topics such as computers, robots and the Nao itself. The main questions we focused on were 1) what a system for answering questions in Dutch and intended for elementary school age children should look like, 2) how the interaction with users should be structured for a smooth interaction and 3) what type of experimental studies and metrics are best for evaluating the quality of the QA system and assess its impact on users. We chose an existing English-language QA system and adapted it to work with Dutch content. An interaction manager was developed for handling the interaction between the system and users. Finally, the system was configured so that it could be interacted with via the Nao robot. The system was then evaluated via two experimental studies. First a quantitative evaluation without users was carried out, evaluating the performance of the system using a collection of test questions. Secondly, a field study with users was done. We took the robot to a day care center where children could ask it questions and assessed the effects of these interactive Q\&A sessions. Results from the quantitative evaluation highlighted slow answering times as the main shortcoming, while recall (70%) and precision (50%) scores were more competitive. The field study showed that even though the occasional repetitiveness and lack of speed were picked up on by users, this did little to curb their enthusiasm. Interaction was smooth, users were interested in many topics the system supported and they continued interacting with it despite the occasional mishap. Overall, the results indicate that, with some speedup, the delivered QA system would be a suitable choice for answering questions about the mentioned topics. ... Read more

Interactive Robotics aims to familiarize children with robotics, which they try to do with their RoboTutor. Towards this goal they developed a platform for controlling the robots but this platform only worked for the Nao robot and not for the Pepper robot.
So for our bachelor end project we have been asked to build a connection tunnel between the server and the Nao and Pepper robots so that it is possible to also use the Pepper robot.

As our research progressed however, it became more clear to us that in order to make a generic solution for both the robots the existing platform would not suffice. So instead of only creating the tunnel we also worked on a new server side solution.
the solution we came up with, virtualization, was sure to solve the problems that caused the Pepper robot to be incompatible with the current platform. By running the server side processes for each robot inside their own environment using Docker containers we where able to create a generic solution for both robots. And with some custom code on the robots themselves, we could ensure that the robots would always connect to the server if they where connected to the Internet. ... Read more

A large problem that primary schools face is that the ratio of pupils to teachers is too high, the class sizes are too large and this makes it difficult for a single teacher to have a good oversight of how the development of a given child is going. The aim of Interactive Robotics is to tackle this problem by bringing robots into the classroom to aid teachers. They aim to have a single robot in a classroom that has the ability to teach different lessons and subjects; the RekenRobot being specifically for basic arithmetic. During the research phase, ideas were gathered regarding how to create teaching methods that are motivating and stimulating. For instance, personalisation, humanising the robot and adaptability of the teaching material were desired functions. The software for the RekenRobot was built from scratch, using the programming language GOAL, JavaScript, CSS, HTML and JSP. The original target audience of the project were children between the ages of 6 and 8. Later this was changed to cover different school years: 3-4, 5-6 and 7-8, making use of levels with different degrees of difficulty. The robot can work one-on-one with a child, being able to practice addition, subtraction, times tables and telling time, as well as 2 forms of explanations can be given: making use of a bus and a number line. Using no explanation to rather focus on automation is also an option. The idea of the project was to lay the groundwork for the later development of the RekenRobot, as this will be an ongoing project for Interactive Robotics. The application designed in this project will be adapted to become part of the Interactive Robotics system. The first user tests at primary schools yielded a largely positive result. The children were excited and motivated to work with the product. The system is simple enough to require very little explanation. This project was never meant to realise a product that can be deployed tomorrow, but the result is a very solid basis for further improvements. ... Read more

A world without digital images is unthinkable in this era of information and communication technology. Billions of images are created, shared and ultimately enjoyed by users every day. However, digital images are sensitive to a wide variety of distortions during the delivery mechanism it goes through. Any of the distortions that can arise during the delivery might disrupt the perceptual quality of an image and can incite dissatisfaction from the user. Thus, it is important to optimize the delivery pipeline towards the arrangement of perceptually good results. For that the perceptual image quality needs to be estimated. Image quality assessment (IQA) is concerned with measuring the degradation in quality of images. The level of degradation needs to be measured in a way that is compliant to how humans would perceive it as humans are the final judges of the delivered quality of an image. General purpose image quality metrics are based on a two-step design. First, quality-aware features are extracted, describing artifact appearance to the Human Visual System. Secondly, these features are mapped to a perceived quality score. General purpose metrics require a large number of varied examples of pairs of distorted images and perceived quality scores to acquire an accurate distortion-agnostic quality prediction. Obtaining a distorted image is relatively simple. However, acquiring the quality assessment is a complex, laborious and expensive undertaking. Consequently, all of the widely known image quality datasets only accommodate at most several thousand examples which is in stark contrast to the mound of possible distortion types and distortion intensities there could be. This limitation hinders the creation of an accurate general-purpose image quality assessment metrics. It would be desirable to extend these datasets with more subjectively evaluated images, but given the cost of doing so, one may want to have a smart mechanism to select those that are most informative to improve the accuracy and robustness of quality metrics. Active learning is such a smart selection mechanism and could therefore be a possible solution. The main goal of this thesis is to explore whether active learning can be beneficial to improve the accuracy of image quality metrics. ... Read more