This Bachelor graduation project has the goal to create a device which is able of automatic volume control, to be used for enhancing speech intelligibility. To tackle the intelligibility of speech through Public Address Systems (PA Systems), an Intelligibility-Enhancing Automatic Volume Control system was proposed. The total system to be made must be able to alter a clean speech signal according to a noise estimation. Then the altered, enhanced, signal should be amplified before being sent to an existing Public Address System. A subsystem is added in order to dampen the outside noise in a car-like environment. The whole project is divided into three parts: Noise Statistics Estimation, Intelligibility Enhancement and Amplifier and Noise Cancellation. These parts have been performed by three different subgroups. In this report, the Amplifier and Noise Cancellation is discussed. The other parts are explained in the respective reports [1,2]. The Amplifier and Noise Cancellation group will amplify the enhanced audio signal with the use of a pre-amplifier. This group also introduces an additional noise cancellation subsystem for usage in enclosed spaces, like a car. It does so by inverting the recorded environment noise below 500 Hz, and adding this to the to be amplified signal before sending it to the PA System. This thesis is divided in two main design sections: the design of the pre-amplifier and the design of the active noise cancellation circuit. At the heart of both circuits lies a LM386 Audio Operational Amplifier (Op-Amp) but they both have different objectives. The pre-amplifier is designed to have a flat transfer function in audio range, 20 Hz to 20 kHz. The output level of the pre-amplifier is a standard level for consumer electronics, being 447 mVpp. The pre-amplifier inverts the signal from the microphone and the audio signal to achieve noise cancellation. The noise cancellation circuit features a microphone amplifier and a Low Pass Filter (LPF). The microphone amplifier amplifies the signal so that the microphone circuit’s output level is at the same level of the audio input of the pre-amplifier (200 mVpp). The filter makes sure only sounds below 500 Hz are passed to the pre-amplifier. With the inverting capabilities of the pre-amplifier and both signals being completely out of phase, a theoretical cancellation of sound signals is possible. Because of the LPF used in the microphone amplifier this cancellation is done for signals below 500 Hz. At the end of the project, a system was built which met most of the requirements. Some of the requirements can not be satisfied due to incapability of the test equipment available. The system does amplify the signal to the desired amplitude and is capable of slightly cancelling noise in a car. However, improvements of the product are needed to function more optimally.

Spatial cues allow a listener to determine the direction sound is coming from. In addition, recognising spatially separated sound sources facilitate the listener to focus on specific sound sources. Because of this, preservation of spatial cues in multi-microphone hearing assistive devices is important to the listening experience and safety of the user.
A number of linearly-constrained-minimum-variance-based methods exist for this purpose. Most of these are limited in the number of interfering sources for which they can preserve the spatial cues. In this thesis, a method of selecting the most important interfering point sources using convex optimisation is proposed.
The method is presented based on two different convex relaxations, which are compared, using simulation experiments, to existing, exhaustive search and randomised methods in terms of noise suppression and localisation errors.
Both methods are shown to improve the performance of the joint binaural linearly constrained minimum variance beamformer, an existing method for simultaneous noise reduction and spatial cue preservation, by giving it more degrees of freedom for noise reduction and allowing it to handle a larger number of (virtual) sources present in the scene.

Authentication is becoming an increasingly important application in the connected world and is driven by the growing use of mobile and IoT devices that use an increasing number of applications that require transactions of sensitive data. Security usually relies on passwords and/or two-factor authentication which are too intrusive for daily use. Biometric solutions such as fingerprints, voice, iris and retina are a good alternative to overcome previous problems. In this project an audio-visual identity verification is presented, where the use of multiple modes that can already be captured from most IoT devices (microphone and camera) make authentication robust to adverse conditions. End-factor analysis (i-vectors) with cosine distance is implemented as the main classification algorithm which takes into account variations within and between speakers. Mel Frequencies Cepstrum Coefficients (MFCC) are used as audio features, 2D-DCT coefficients of a single snapshot and Motion Vectors (MV) of the lips are extracted for visual features. Improvements combining different modes are shown using VidTimit dataset where the proposed algorithm achieves 0.7% of Half Total Error (HTER) in the test set outperforming single modes audio and visual by 9.5% and 6.4%, respectively.