Nowadays, speech synthesis is part of various daily life applications. The ultimate goal of such technologies consists in extending the possibilities of interaction with the machine, in order to get closer to human-like communications. However, current state-of-the-art systems often lack of realism: although high-quality speech synthesis can be produced by many researchers and companies around the world, synthetic voices are generally perceived as hyperarticulated. In any case, their degree of articulation is fixed once and for all. The present thesis falls within the more general quest for enriching expressivity in speech synthesis. The main idea consists in improving statistical parametric speech synthesis, whose most famous example is Hidden Markov Model (HMM) based speech synthesis, by introducing a control of the articulation degree, so as to enable synthesizers to automatically adapt their way of speaking to the contextual situation, like humans ...

This thesis presents a novel approach to speech enhancement by exploiting the bimodality of speech production and the correlation that exists between audio and visual speech information. An analysis into the correlation of a range of audio and visual features reveals significant correlation to exist between visual speech features and audio filterbank features. The amount of correlation was also found to be greater when the correlation is analysed with individual phonemes rather than across all phonemes. This led to building a Gaussian Mixture Model (GMM) that is capable of estimating filterbank features from visual features. Phoneme-specific GMMs gave lower filterbank estimation errors and a phoneme transcription is decoded using audio-visual Hidden Markov Model (HMM). Clean filterbank estimates along with mean noise estimates were then utilised to construct visually-derived Wiener filters that are able to enhance noisy speech. The mean noise ...

In the world of Information and communications technology (ICT), strategies for innovation and development are increasingly focusing on applications that require spatial representation and real-time interaction with and within 3D-media environments. One of the major challenges that such applications have to address is user-centricity, reflecting e.g. on developing complexity-hiding services so that people can personalize their own delivery of services. In these terms, multimodal interfaces represent a key factor for enabling an inclusive use of new technologies by everyone. In order to achieve this, multimodal realistic models that describe our environment are needed, and in particular models that accurately describe the acoustics of the environment and communication through the auditory modality are required. Examples of currently active research directions and application areas include 3DTV and future internet, 3D visual-sound scene coding, transmission and reconstruction and teleconferencing systems, to name but ...

Spatial hearing for human listeners is based on the interaural as well as on the monaural analysis of the signals arriving at both ears, enabling the listeners to assign certain spatial components to these signals. This spatial aspect gets lost when the signals are reproduced via headphones without considering the acoustical influence of the head and torso, i.e. head-related transfer function (HRTFs). A common procedure to take into account spatial aspects in a binaural reproduction is to use so-called artificial heads. Artificial heads are replicas of a human head and torso with average anthropometric geometries and built-in microphones in the ears. Although, the signals recorded with artificial heads contain relevant spatial aspects, binaural recordings using artificial heads often suffer from front-back confusions and the perception of the sound source being inside the head (internalization). These shortcomings can be attributed to ...

The narrowband frequency range of telephone speech signals originally caused by former analog transmission techniques still leads to frequent acoustical limitations in today’s digital telephony systems. It provokes muffled sounding phone calls with reduced speech intelligibility and quality. By means of artificial speech bandwidth extension approaches, missing frequency components can be estimated and reconstructed. However, the artificially extended speech bandwidth typically suffers from annoying artifacts. Particularly susceptible to this are the fricatives /s/ and /z/. They can hardly be estimated based on the narrowband spectrum and are therefore easily confusable with other phonemes as well as speech pauses. This work takes advantage of phonetic a priori knowledge to optimize the performance of artificial bandwidth extension. Both the offline training part conducted in advance and the main processing part performed later on shall be thereby provided with important phoneme information. As ...

This dissertation introduces interactive designs in the context of statistical parametric synthesis. The objective is to develop methods and designs that enrich the Human-Computer Interaction by enabling computers (or other devices) to have more expressive and adjustable voices. First, we tackle the problem of interactive controls and present a novel method for performative HMM-based synthesis (pHTS). Second, we apply interpolation methods, initially developed for the traditional HMM-based speech synthesis system, in the interactive framework of pHTS. Third, we integrate articulatory control in our interactive approach. Fourth, we present a collection of interactive applications based on our work. Finally, we unify our research into an open source library, Mage. To our current knowledge Mage is the first system for interactive programming of HMM-based synthesis that allows realtime manipulation of all speech production levels. It has been used also in cases that ...

In digital speech-communication systems like mobile phones, public address systems and hearing aids, conveying the message is one of the most important goals. This can be challenging since the intelligibility of the speech may be harmed at various stages before, during and after the transmission process from sender to receiver. Causes which create such adverse conditions include background noise, an unreliable internet connection during a Skype conversation or a hearing impairment of the receiver. To overcome this, many speech-communication systems include speech processing algorithms to compensate for these signal degradations like noise reduction. To determine the effect on speech intelligibility of these signal processing based solutions, the speech signal has to be evaluated by means of a listening test with human listeners. However, such tests are costly and time consuming. As an alternative, reliable and fast machine-driven intelligibility predictors are ...

This thesis faces the problem of automatically classifying environmental sounds, i.e., any non-speech or non-music sounds that can be found in the environment. Broadly speaking, two main processes are needed to perform such classification: the signal feature extraction so as to compose representative sound patterns and the machine learning technique that performs the classification of such patterns. The main focus of this research is put on the former, studying relevant signal features that optimally represent the sound characteristics since, according to several references, it is a key issue to attain a robust recognition. This type of audio signals holds many differences with speech or music signals, thus specific features should be determined and adapted to their own characteristics. In this sense, new signal features, inspired by the human auditory system and the human perception of sound, are proposed to improve ...

Cross-lingual voice conversion refers to the automatic transformation of a source speaker’s voice to a target speaker’s voice in a language that the target speaker can not speak. It involves a set of statistical analysis, pattern recognition, machine learning, and signal processing techniques. This study focuses on the problems related to cross-lingual voice conversion by discussing open research questions, presenting new methods, and performing comparisons with the state-of-the-art techniques. In the training stage, a Phonetic Hidden Markov Model based automatic segmentation and alignment method is developed for cross-lingual applications which support textindependent and text-dependent modes. Vocal tract transformation function is estimated using weighted speech frame mapping in more detail. Adjusting the weights, similarity to target voice and output quality can be balanced depending on the requirements of the cross- lingual voice conversion application. A context-matching algorithm is developed to reduce ...

People of all generations are making more and more use of digital imaging systems in their daily lives. The image content rendered by these digital imaging systems largely differs in perceived quality depending on the system and its applications. To be able to optimize the experience of viewers of this content understanding and modeling perceived image quality is essential. Research on modeling image quality in a full-reference framework --- where the original content can be used as a reference --- is well established in literature. In many current applications, however, the perceived image quality needs to be modeled in a no-reference framework at real-time. As a consequence, the model needs to quantitatively predict perceived quality of a degraded image without being able to compare it to its original version, and has to achieve this with limited computational complexity in order ...

When listening to music, some humans can easily recognize which instruments play at what time or when a new musical segment starts, but cannot describe exactly how they do this. To automatically describe particular aspects of a music piece – be it for an academic interest in emulating human perception, or for practical applications –, we can thus not directly replicate the steps taken by a human. We can, however, exploit that humans can easily annotate examples, and optimize a generic function to reproduce these annotations. In this thesis, I explore solving different music perception tasks with deep learning, a recent branch of machine learning that optimizes functions of many stacked nonlinear operations – referred to as deep neural networks – and promises to obtain better results or require less domain knowledge than more traditional techniques. In particular, I employ ...

Reverberation is the sum of reflected sound waves and is present in any conventional room. Speech communication devices such as mobile phones in hands-free mode, tablets, smart TVs, teleconferencing systems, hearing aids, voice-controlled systems, etc. use one or more microphones to pick up the desired speech signals. When the microphones are not in the proximity of the desired source, strong reverberation and noise can degrade the signal quality at the microphones and can impair the intelligibility and the performance of automatic speech recognizers. Therefore, it is a highly demanded task to process the microphone signals such that reverberation and noise are reduced. The process of reducing or removing reverberation from recorded signals is called dereverberation. As dereverberation is usually a completely blind problem, where the only available information are the microphone signals, and as the acoustic scenario can be non-stationary, ...

Auditory modeling is a well-established methodology that provides insight into human perception and that facilitates the extraction of signal features most relevant to the human listener for coding applications. This thesis deals with the approach of `coding in the perceptual domain' and is based on an invertible auditory model that provides a pulsed auditory representation of the input speech or audio signal. It is natural for pulsed signal representations to encode only the non-zero samples by specifying their positions as side information. For the considered auditory representation, the number of pulses and, therefore, the amount of side information is too high for an efficient encoding at a relatively low bit rate. The focus of this work is to `sparsify' the pulsed signal representation, i.e., to remove its perceptual irrelevance and its redundancy, to obtain a compact signal representation, which facilitates ...

Due to their decreased ability to understand speech hearing impaired may have difficulties to interact in social groups, especially when several people are talking simultaneously. Fortunately, in the last decades hearing aids have evolved from simple sound amplifiers to modern digital devices with complex functionalities including noise reduction algorithms, which are crucial to improve speech understanding in background noise for hearing-impaired persons. Since many hearing aid users are fitted with two hearing aids, so-called binaural hearing aids have been developed, which exchange data and signals through a wireless link such that the processing in both hearing aids can be synchronized. In addition to reducing noise and limiting speech distortion, another important objective of noise reduction algorithms in binaural hearing aids is the preservation of the listener’s impression of the acoustical scene, in order to exploit the binaural hearing advantage and ...

Optimizing the performance of digital imaging systems with respect to the capture, display, storage and transmission of visual information represents one of the biggest challenges in the field of image and video processing. Taking into account the way humans perceive visual information can be greatly beneficial for this task. To achieve this, it is necessary to understand and model the human visual system, which is also the principal goal of this thesis. Computational models for different aspects of the visual system are developed, which can be used in a wide variety of image and video processing applications. The proposed models and metrics are shown to be consistent with human perception. The focus of this work is visual quality assessment. A perceptual distortion metric (PDM) for the evaluation of video quality is presented. It is based on a model of the ...