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, ...

Reducing interference noise in a noisy speech recording has been a challenging task for many years yet has a variety of applications, for example, in handsfree mobile communications, in speech recognition, and in hearing aids. Traditional single-channel noise reduction schemes, such as Wiener filtering, do not work satisfactorily in the presence of non-stationary background noise. Alternatively, supervised approaches, where the noise type is known in advance, lead to higher-quality enhanced speech signals. This dissertation proposes supervised and unsupervised single-channel noise reduction algorithms. We consider two classes of methods for this purpose: approaches based on nonnegative matrix factorization (NMF) and methods based on hidden Markov models (HMM). The contributions of this dissertation can be divided into three main (overlapping) parts. First, we propose NMF-based enhancement approaches that use temporal dependencies of the speech signals. In a standard NMF, the important temporal ...

Audio systems receive the speech signals of interest usually in the presence of noise. The noise has profound impacts on the quality and intelligibility of the speech signals, and it is therefore clear that the noisy signals must be cleaned up before being played back, stored, or analyzed. We can estimate the speech signal of interest from the noisy signals using a priori knowledge about it. A human speech signal is broadband and consists of both voiced and unvoiced parts. The voiced part is quasi-periodic with a time-varying fundamental frequency (or pitch as it is commonly referred to). We consider the periodic signals basically as the sum of harmonics. Therefore, we can pass the noisy signals through bandpass filters centered at the frequencies of the harmonics to enhance the signal. In addition, although the frequencies of the harmonics are the ...

In an automotive environment, quality of speech communication using a hands-free equipment is often deteriorated by interfering car noise. In order to preserve the speech signal without car noise, a multichannel speech enhancement system including a beamformer and a post-filter can be applied. Since employing a beamformer alone is insufficient to substantially reducing the level of car noise, a post-filter has to be applied to provide further noise reduction, especially at low frequencies. In this thesis, two novel post-filter designs along with their optimization for different driving conditions are presented. The first post-filter design utilizes an adaptive smoothing factor for the power spectral density estimation as well as a hybrid noise coherence function. The hybrid noise coherence function is a mixture of the diffuse and the measured noise coherence functions for a specific driving condition. The second post-filter design applies ...

Never before has speech been captured as often by electronic devices equipped with one or multiple microphones, serving a variety of applications. It is the key aspect in digital telephony, hearing devices, and voice-driven human-to-machine interaction. When speech is recorded, the microphones also capture a variety of further, undesired sound components due to adverse acoustic conditions. Interfering speech, background noise and reverberation, i.e. the persistence of sound in a room after excitation caused by a multitude of reflections on the room enclosure, are detrimental to the quality and intelligibility of target speech as well as the performance of automatic speech recognition. Hence, speech enhancement aiming at estimating the early target-speech component, which contains the direct component and early reflections, is crucial to nearly all speech-related applications presently available. In this thesis, we compare, propose and evaluate existing and novel approaches ...

Speech communication frequently suffers from transmitted background noises. Numerous speech enhancement algorithms have thus been proposed to obtain a speech signal with a reduced amount of background noise and better speech quality. In most cases they are analytically derived as spectral weighting rules for given error criteria along with statistical models of the speech and noise spectra. However, as these spectral distributions are indeed not easy to be measured and modeled, such algorithms achieve in practice only a suboptimal performance. In the development of state-of-the-art algorithms, speech and noise training data is commonly exploited for the statistical modeling of the respective spectral distributions. In this thesis, the training data is directly applied to train data-driven speech enhancement algorithms, avoiding any modeling of the spectral distributions. Two applications are proposed: (1) A set of spectral weighting rules is trained from noise ...

This thesis describes the application of a perceptually based spectral subtraction algorithm for the enhancement of non-stationary noise corrupted speech. Through examination of speech enhancement techniques, explanations are given for the choice of magnitude spectral subtraction and how the human auditory system can be modelled for frequency domain speech enhancement. It is discovered, that the cochlea provides the mechanical speech enhancement in the auditory system, through the use of masking. Frequency masking is used in spectral subtraction, to improve the algorithm execution time, and to shape the enhancement process making it sound natural to the ear. A new technique for estimation of background noise is presented, which operates during speech sections as well as pauses. This uses two microphones placed on opposite ends of the cellular handset. Using these, the algorithm determines whether the signal is speech, or noise, by ...

The research in this thesis addresses the problem of automatic speech recognition in noisy environments. Automatic speech recognition systems obtain acceptable performances in noise free conditions but these performances degrade dramatically in presence of additive noise. This is mainly due to the mismatch between the training and the noisy operating conditions. In the time-frequency representation of the noisy speech signal, some of the clean speech features are masked by noise. In this case the clean speech features cannot be correctly estimated from the noisy speech and therefore they are considered as missing or unreliable. In order to improve the performance of speech recognition systems in additive noise conditions, special attention should be paid to the problems of detection and compensation of these unreliable features. This thesis is concerned with the problem of missing features applied to automatic speaker-independent speech recognition. ...

According to the Parkinson’s Foundation, more than 10 million people world- wide suffer from Parkinson’s disease (PD). The common symptoms are tremor, muscle rigidity and slowness of movement. There is no cure available cur- rently, but clinical intervention can help alleviate the symptoms significantly. Recently, it has been found that PD can be detected and telemonitored by voice signals, such as sustained phonation /a/. However, the voiced-based PD detector suffers from severe performance degradation in adverse envi- ronments, such as noise, reverberation and nonlinear distortion, which are common in uncontrolled settings. In this thesis, we focus on deriving speech modeling and robust estima- tion algorithms capable of improving the PD detection accuracy in adverse environments. Robust estimation algorithms using parametric modeling of voice signals are proposed. We present both segment-wise and sample-wise robust pitch tracking algorithms using the harmonic model. ...

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 this thesis, we present novel methods to enhance speech corrupted by noise. All methods are based on the processing of complex-valued spectral data. First, kernel principal component analysis (PCA) for speech enhancement is proposed. Subsequently, a simplification of kernel PCA, called pre-image iterations (PI), is derived. This method computes enhanced feature vectors iteratively by linear combination of noisy feature vectors. The weighting for the linear combination is found by a kernel function that measures the similarity between the feature vectors. The kernel variance is a key parameter for the degree of de-noising and has to be set according to the signal-to-noise ratio (SNR). Initially, PI were proposed for speech corrupted by additive white Gaussian noise. To be independent of knowledge about the SNR and to generalize to other stationary noise types, PI are extended by automatic determination of the ...

This thesis addresses the problem of multichannel audio source separation by exploiting deep neural networks (DNNs). We build upon the classical expectation-maximization (EM) based source separation framework employing a multichannel Gaussian model, in which the sources are characterized by their power spectral densities and their source spatial covariance matrices. We explore and optimize the use of DNNs for estimating these spectral and spatial parameters. Employing the estimated source parameters, we then derive a time-varying multichannel Wiener filter for the separation of each source. We extensively study the impact of various design choices for the spectral and spatial DNNs. We consider different cost functions, time-frequency representations, architectures, and training data sizes. Those cost functions notably include a newly proposed task-oriented signal-to-distortion ratio cost function for spectral DNNs. Furthermore, we present a weighted spatial parameter estimation formula, which generalizes the corresponding exact ...

The topic of this thesis is methods of pre-processing speech signals for robust estimation of model parameters in models of these signals. Here, there is a special focus on the situation where the desired signal is contaminated by colored noise. In order to estimate the speech signal, or its voiced and unvoiced components, from a noisy observation, it is important to have robust estimators that can handle colored and non-stationary noise. Two important aspects are investigated. The first one is a robust estimation of the speech signal parameters, such as the fundamental frequency, which is required in many contexts. For this purpose, fast estimation methods based on a simple white Gaussian noise (WGN) assumption are often used. To keep using those methods, the noisy signal can be pre-processed using a filter. If the colored noise is modelled as an autoregressive ...

Multiple pitch tracking of speech is an important task for the segregation of multiple speakers in a single-channel recording. In this thesis, a probabilistic model-based approach for estimation and tracking of multiple pitch trajectories is proposed. A probabilistic model that captures pitch-dependent characteristics of the single-speaker short-time spectrum is obtained a priori from clean speech data. The resulting speaker model, which is based on Gaussian mixture models, can be trained either in a speaker independent (SI) or a speaker dependent (SD) fashion. Speaker models are then combined using an interaction model to obtain a probabilistic description of the observed speech mixture. A factorial hidden Markov model is applied for tracking the pitch trajectories of multiple speakers over time. The probabilistic model-based approach is capable to explicitly incorporate timbral information and all associated uncertainties of spectral structure into the model. While ...

In modern devices which provide hands-free speech capturing functionality, such as hands-free communication kits and voice-controlled devices, the received speech signal at the microphones is corrupted by background noise, interfering speech signals, and room reverberation. In many practical situations, the microphones are not necessarily located near the desired source, and hence, the ratio of the desired speech power to the power of the background noise, the interfering speech, and the reverberation at the microphones can be very low, often around or even below 0 dB. In such situations, the comfort of human-to-human communication, as well as the accuracy of automatic speech recognisers for voice-controlled applications can be signi cantly degraded. Therefore, e ffective speech enhancement algorithms are required to process the microphone signals before transmitting them to the far-end side for communication, or before feeding them into a speech recognition ...