Results 1 - 10 of 158
Results 1 - 10 of 158. Search took: 0.021 seconds
|Sort by: date | relevance|
[en] Very low power, short-range microwave ''radar-like'' sensors can measure the motions and vibrations of internal human speech articulators as speech is produced. In these animate (and also in inanimate acoustic systems) microwave sensors can measure vibration information associated with excitation sources and other interfaces. These data, together with the corresponding acoustic data, enable the calculation of system transfer functions. This information appears to be useful for a surprisingly wide range of applications such as speech coding and recognition, speaker or object identification, speech and musical instrument synthesis, noise cancellation, and other applications
[en] Uvulopalatopharyngoplasty (UPPP) is a commonly used surgical technique for oropharyngeal reconstruction in patients with obstructive sleep apnea (OSA). This procedure can be done either through the classic or the laser-assisted uvulopalatopharyngoplasty (LAUP) technique. The purpose of this study was to evaluate the effect of classic UPPP and LAUP on acoustics of voice and speech nasalance, and to compare the effect of each operation on these two domains. Patients and The study included 27 patients with a mean age of 46 years. All patients were diagnosed with OSA based on polysomnographic examination. Patients were divided into two groups according to the type of surgical procedure. Fifteen patients underwent classic UPPP, whereas 12 patients were subjected to LAUP. A full assessment was done for all patients preoperatively and postoperatively, including auditory perceptual assessment (APA) of voice and speech, objective assessment using acoustic voice analysis and nasometry. Auditory perceptual assessment of speech and voice, acoustic analysis of voice and nasometric analysis of speech did not show statistically significant differences between the preoperative and postoperative evaluations in either group (P>.05).The results of this study demonstrated that in patients with OSA, the surgical technique, whether classic UPPP or LAUP, does not have significant effects on the patients' voice quality or their speech outcomes (Author).
[en] The present paper deals with the real-time speech synthesis implemented on a minicomputer. A first program translates the orthographic text into a string of phonetic codes, which is then processed by the synthesis program itself. The method used, a synthesis by rules, directly computes the speech signal in its amplitude-time representation. Emphasis has been put on special cases (diphthongs, 'e muet', consonant-consonant transition) and the implementation of the rhythm and of the melody. (author)
[fr]L'objet de ce memoire est la synthese de la parole en temps reel au moyen d'un mini-ordinateur. Un premier programme traduit le texte orthographique en une chaine de codes phonetiques qui est traitee ensuite par le programme de synthese proprement dit. La methode de synthese par regles utilisee calcule directement le signal de parole dans sa representation amplitude-temps. Nous avons etudie specialement certains cas particuliers (diphtongues, e muet, transition consonne-consonne) et la mise en oeuvre du rythme et de la melodie. (auteur)
[en] Acoustical analysis of speech using computers has reached an important development in the latest years. The subjective evaluation of a clinician is complemented with an objective measure of relevant parameters of voice. Praat, MDVP (Multi Dimensional Voice Program) and SAV (Software for Voice Analysis) are some examples of software for speech analysis. This paper describes an approach to estimate the subjective characteristics of RASATI scale given objective acoustical parameters. Two approaches were used: linear regression with non-negativity constraints, and neural networks. The experiments show that such approach gives correct evaluations with ±1 error in 80% of the cases.
[en] We present a new algorithm for bound-constrained total-variation (TV) regularization that in comparison with its predecessors is simple, fast, and flexible. We use a splitting approach to decouple TV minimization from enforcing the constraints. Consequently, existing TV solvers can be employed with minimal alteration. This also makes the approach straightforward to generalize to any situation where TV can be applied. We consider deblurring of images with Gaussian or salt-and-pepper noise, as well as Abel inversion of radiographs with Poisson noise. We incorporate previous iterative reweighting algorithms to solve the TV portion.
[en] The use of EM radiation in conjunction with simultaneously recorded acoustic speech information enables a complete mathematical coding of acoustic speech. The methods include the forming of a feature vector for each pitch period of voiced speech and the forming of feature vectors for each time frame of unvoiced, as well as for combined voiced and unvoiced speech. The methods include how to deconvolve the speech excitation function from the acoustic speech output to describe the transfer function each time frame. The formation of feature vectors defining all acoustic speech units over well defined time frames can be used for purposes of speech coding, speech compression, speaker identification, language-of-speech identification, speech recognition, speech synthesis, speech translation, speech telephony, and speech teaching. 35 figs
[en] CERN celebrated its 40th anniversary with a special family day on 17 September where the accent was on fun, games and entertainment for staff and their families. After a brief opening ceremony, pomp gave way to play, formality to frivolity. The following extracts from the DG's speech of welcome were freely translated from the French original: 'Forty is the traditional age for taking stock of the past and assessing future prospects. CERN is no exception to this rule and it is particularly appropriate at what is clearly a major turning point in its existence. Before looking to the future, allow me to say a few words about our past
[en] This paper deals with problems of speech transmission quality measurement in modern telecommunication networks. It focuses on problems caused by specific types of distortions and errors caused present in transmissions using TCP/IP networks
[en] An algorithm is presented for the calculation of the prosodic parameters for speech synthesis. It uses the melodic patterns, composed of rising and falling slopes, suggested by G. CAELEN, and rests on: 1. An analysis into units of meaning to determine a melodic pattern 2. the calculation of the numeric values for the prosodic variations of each syllable; 3. The use of a table of vocalic values for the three parameters for each vowel according to the consonantal environment and of a table of standard duration for consonants. This method was applied in the 'SARA' program of synthesis with satisfactory results. (author)
[fr]Ce memoire presente une methode de calcul automatique des valeurs prosodiques (frequence fondamentale, duree, energie) des phonemes d'une phrase pour la synthese de la parole. Cette methode utilise les schemas prosodiques de G. CAELEN. Ces schemas sont composes de pentes croissantes et decroissantes. Ces pentes peuvent etre representees par des valeurs numeriques qui rendent compte des variations prosodiques dans la phrase. Cette etude comporte trois etapes: 1. Un decoupage de la phrase en groupes de sens; ceci permet le calcul d'un schema prosodique. 2. Le calcul des valeurs numeriques donnant pour chaque syllabe de la phrase les variations de la prosodie. 3. L'utilisation: - d'un tableau de valeurs vocaliques resultantes qui representent la duree, l'energie et la frequence fondamentale standard de chaque voyelle, en fonction d'un contexte consonantique; - d'un tableau de duree standard pour les consonnes. Nous decrivons l'algorithme qui permet d'obtenir les valeurs prosodiques a partir des valeurs numeriques, des valeurs vocaliques resultantes et des durees standard pour les consonnes. Les valeurs obtenues ont ete appliquees au programme de synthese 'SARA'. Nous observons une amelioration de l'intelligibilite due a une meilleure prosodie. (auteur)