Our research focuses on speech detection from body movements using wearable accelerometer data collected in an in-the-wild mingling event. We aim to explore the nature of the connection between speech and body movements. More specifically, we stress on the person-specificity of speech. Many studies have shown that speech always comes along with unconscious body behaviours. There is a strong correlation and synchrony between speech and body movements. Previous research has proved that human behaviour is highly person-specific. In other words, in our experiment set- up, the accelerometer data distributions collected from different persons are different. Based on the two considerations discussed above, our work contains two phases. In the first phase, we investigate utilizing convolutional and recurrent neural networks for learning informative representations from raw body acceleration readings. The model we proposed outperforms the state-of-the-art approach presented in [5] by 6 % (Area Under the Curve) with the same data. In the next stage, we visualize the features extracted by the proposed model. The results show that distributions of data obtained from different individuals can differ (also known as person-specificity of the problem). We adopt two approaches of multi-source domain adaption [6] based on the features extracted by our model, aiming to form a personalized speech detection model for each person in our dataset. The first approach is called transductive parameter transfer (TPT) [5]. It deduces the personalized model of the target domain from the known well-trained models of several source domains based on the assumption that distributions of individuals with similar marginal distributions should also have similar decision boundaries. The second strategy is a sample re-weighting based method where the training samples from different persons are re-weighted with respect to the similarities of their conditional and marginal distributions to the target person. We use those re-weighted samples to train a personalized model for each target person. The approaches we adopted only achieved a relative performance increase compared to the general neural network model trained on all the data. We then discuss the possible reasons why these two methods did not bring significant improvement and what can be the alternative solution in the future.

Sick Building Syndrome is present in 30% of all office buildings and can cause serious health damage over time. This is an era where sustainability and well-being are becoming dominant aspects of life. As a result, it is becoming increasingly important to businesses to invest in their employees' well-being and health. The VTTI group cares for the well-being of their employees, and is looking for a tool to optimize the utilization of their building for perceived thermal comfort and indoor air quality.

This report documents the development of Claire, an indoor air quality dashboard that helps to identify local air quality problems. Using Claire, employees can be rearranged throughout the space, learn about the characteristics of their office, and for example switch to another meeting room. Claire translates measurements into insights. Claire learns about the behavior of the office, and gives recommendations once she notices that the indoor air quality can be improved.

Claire is backed by an indoor air quality sensor mesh network, which has been developed as part of this project. The sensors continuously measure temperature, humidity and carbon dioxide concentrations. The sensors connect to a cloud infrastructure through a local internet gateway. In the cloud the data gets processed. All measurements are displayed real-time in the dashboard.

Claire is different from existing products in several ways. First, the sensors developed measure both dry-bulb and black globe temperature, which gives it a temperature reading that describes human thermal comfort more accurately. This is not done in competing products. Furthermore, the sensors fill the gap for small and medium-sized enterprises (SMEs): the sensor network is able to get fine-grained results due to its high sensor density, whilst still being very easy to setup with no adjustments to the building being required. Finally, the developed data analysis methods translate the measurements from the sensor network to concrete suggestions, sent through a push notification, which enables workers to get involved with improving the indoor air quality in their office space.