Auswahl der wissenschaftlichen Literatur zum Thema „Glottal excitation source“

Now a days everyone is using mobile phones. Mobile phones play an important role in communication. While using mobile phones, people face the problem of noise signals. These noise signals are affecting the quality of speech signals. These noise signals maybe produced from the materials or sources such as echoes, crowded places, etc. For example, when we speak over the mobile phone or in any crowded place, the external noise adds up like the noise of the subway, train, car, etc. In real time to remove external noise is a very crucial task. A lot of research has been done in cancelling the background noise reduction. Acoustic noise is a type of noise which can produce unwanted sound and it can be reduced by selecting a higher switching frequency (at the cost of higher inverter losses). So a new technique was introduced called passive noise cancellation that supresses higher frequency acoustic noise. In order to supress the lower frequency, the passive techniques require materials that is too bulky and heavy. For these materials, an alternative method is required known as active noise cancellation. This technique does not come without its change as it is used to actively cancel the noise of the world around you to make your audio come clearer. Active noise cancellation separates noise signal and the speech signal is chosen. In order to use active noise cancellation, we use zero frequency filter technique for the cancellation of noisy speech signals. It is used for detecting the regions of glottal activity and in estimating the strength of excitation n each glottal cycle. The main advantage of active noise cancellation is to cancel the random sounds due to repetition in the waveform. ANC is used in the reduction by using a power source. It is best suited for low frequencies. VLSI architecture for zero frequency filter can be used as a voice processor in mobile applications. The existing system has produced a time delay of 220ps. In this paper, we are reducing the time delay into 120ps by using tanner tool software.

Abstract Introduction People with Parkinson’s Disease (PD) have difficult motor hearing calibration, which affects the monitoring of vocal production. Therefore, employing visual stimulus associated with vocal therapy may benefit the voice of this population. For this purpose, some techniques favour the competence of the glottis, such as pushing, which consists of movements of the arms with a simultaneous effort to phonation and, associated with virtual reality games, provide playful therapeutic intervention through the simulation of experiences that can foster glottal adjustments and generate changes in the voice. Objectives To determine the immediate effect of the technique of pushing with plosive sounds associated with the use of a virtual reality game in the voice of people with Parkinson's disease. Methodology The participants were 17 men and 14 women, mean age of 62.61 (± 11.24) years, meantime of diagnosis 6.93 years, distributed in stages I (32.5%), II (22.59%), III (45.16%) of Hoehn and Yahr. scale. The vocal evaluation pre and post pushing technique with plosive sounds associated with virtual reality game consisted of analysing vocal, acoustic parameters from the record of the issuance of the vowel /ε/. The parameters extracted were Jitter, Shimmer, corresponding to signs of disturbance of the sound source and Glottal-to Noise Excitation (GNE), corresponding to the energy of the glottis on the noise. The satisfaction questionnaire on the implementation of this technique associated with virtual reality was used. Results The values of disturbance of the sound wave decreased, and the glottal energy on the noise increased, respectively, Jitter (p = 0.0075); Shimmer (p = 0.0048); GNE (p = 0.0008). Regarding the satisfaction questionnaire, 74.20% of them felt very satisfied and 25.80%, satisfied. Conclusion The vocal technique associated with virtual reality game improved the acoustic parameters of the participants' voice and the majority felt great satisfaction with the intervention.

AbstractThis paper presents a proposal to a source-filter theory of voice production, more precisely related to voiced sounds. It is a proposal of a model to generate signal using linear and time-invariant systems and takes into account the phonation biophysics and the cyclostationary characteristics of the voice signal, related to the vibrational behavior of the vocal cords. The model suggests that the oscillation frequency of the vocal cords is a function of its mass and length, but controlled by the longitudinal tension applied to them. The mathematical description of the model of glottal excitation is presented, along with a mathematical closed expression for the power spectral density of the signal that excites the glottis. The voice signal, whose parameters can be adjusted for detection and classification of glottis pathologies, is also present. As a result, the output of each block diagram that represents the proposed model is analysed, including a power spectral density comparison between emulated voice, original voice, and classic source-filter model. The Log Spectral Distortion is computed, providing values below 1.40 dB, indicating an acceptable distortion for all cases.

La Conversion de la Voix (VC) vise à transformer les caractéristiques de la voix d’un locuteur source de manière qu’il sera perçu comme étant prononcé par un locuteur cible. Le principe de la VC est de définir des fonctions du transposition pour la conversion de la voix de l’un locuteur source à la voix de l’un locuteur cible. Les fonctions de transformation de VC systèmes "State-Of-The-Art" (START) adapte instantanément aux caractéristiques de la voix source. Cependant, la qualité est pas encore suffisant. Des améliorations considérables sont nécessaires que les techniques VC peuvent être utilisés dans un environnement industriel professionnel. L’objectif de cette thèse est d’augmenter la qualité de la conversion de la voix pour faciliter son applicabilité industrielle dans une mesure raisonnable. Les propriétés de base de différentes START algorithmes de la conversion de la voix sont discutés sur leurs avantages intrinsèques et ses déficits. Basé sur des évaluations expérimentales avec un GMM VC système la conclusion est que la plupart des systèmes VC START qui reposent sur des modèles statistiques sont, en raison de l’effet en moyenne de la régression linéaire, moins appropriées pour atteindre un score du similitude assez élevé avec le haut-parleur cible requise pour l’utilisation industrielle. Les contributions établies pendant de ce travail de thèse se trouvent dans les moyens étendus à a) modéliser l’excitation du source glottique, b) modéliser des descripteurs de la voix en utilisant un nouveau système de parole basée sur un modèle élargie de source-filtre, et c) avancer une nouveau système VC de l’Ircam en le combinant avec les contributions de a) et b)
Voice Conversion (VC) aims at transforming the characteristics of a source speaker’s voice in such a way that it will be perceived as being uttered by a target speaker. The principle of VC is to define mapping functions for the conversion from one source speaker’s voice to one target speaker’s voice. The transformation functions of common State-Of-The-Art (START) VC system adapt instantaneously to the characteristics of the source voice. While recent VC systems have made considerable progress over the conversion quality of initial approaches, the quality is nevertheless not yet sufficient. Considerable improvements are required before VC techniques can be used in an professional industrial environment. The objective of this thesis is to augment the quality of Voice Conversion to facilitate its industrial applicability to a reasonable extent. The basic properties of different START algorithms for Voice Conversion are discussed on their intrinsic advantages and shortcomings. Based on experimental evaluations of one GMM-based State-Of-The-Art VC approach the conclusion is that most VC systems which rely on statistical models are, due to averaging effect of the linear regression, less appropriate to achieve a high enough similarity score to the target speaker required for industrial usage. The contributions established throughout this thesis work lie in the extended means to a) model the glottal excitation source, b) model a voice descriptor set using a novel speech system based on an extended source-filter model, and c) to further advance IRCAM’s novel VC system by combining it with the contributions of a) and b)