In this paper, we propose parameter generation methods using rich context models in HMM-based speech synthesis as yet another hybrid method combining HMM-based speech synthesis and unit selection synthesis. In the traditional HMM-based speech synthesis, generated speech parameters tend to be excessively smoothed and they cause muffled sounds in synthetic speech. To alleviate this problem, sever...

This research work highlights the effects of acoustic emission (AE) signals emitted during the milling of H13 tool steel as an important parameter in the identification of tool wear. These generated AE signals provide information on the chip formation, wear, fracture and general deformation. Furthermore, it is aimed at implementing an online monitoring system for machine tools, using a sensor f...

Water detection in hydrocarbon emulsions represents a serious challenge in petroleum production industry. In this paper, a systematic and an applicable method of water detection in hydrocarbon emulsion is achieved using thermoacoustic effect. The study focused on sensing the presence of water in kerosene using the established correlation between the content of hydrocarbon emulsion and the chara...

The circular double-layer array of loudspeakers has been proven feasible in reproducing an expected sound field inside the array and reducing sound waves propagating outside the array. Usually the performance is evaluated by two criteria concerning the acoustic pressure matching and the acoustic contrast. In this paper, the standard parameter related to the localization cues of human auditory s...

An entropy-based feature parameter weighting scheme was proposed previously [1], in which the scores obtained from different feature parameters are weighted differently in the decoding process according to an entropy measure. In this paper, we propose a more delicate entropy measure for this purpose considering the inherent confusion among different acoustic classes. If a set of acoustic classe...

ABSTRACT. We study the classical combined field integral equation formulations for time-harmonic acoustic scattering by a sound soft bounded obstacle, namely the indirect formulation due to Brakhage-Werner/Leis/Panič, and the direct formulation associated with the names of Burton and Miller. We obtain lower and upper bounds on the condition numbers for these formulations, emphasising dependence...

The development of simple automatic diagnostic and assessment procedures for the comparison and optimisation of TES-based DVI systems is presented. The use of a "Diagnostic Matrix" [1] to assess the variability of input acoustic events purporting to be the same word and to indicate the degree of orthogonality between the acoustic events which form the vocabulary under examination, is discussed....

The acoustic-phonetic properties of words spoken with three different levels of accentuation (de-accented, pre-nuclear and nuclear accented in broad-focus and nuclear accented in narrow-focus) are examined in question-answer elicited sentences and iterative imitations (on the syllable da) produced by six French and six German speakers. Normalised parameter values allow a comparative weighting o...

This study deals with acoustic phonetic modelling in HMM based continuous speech recognition. Context dependent phone models were derived by a decision tree clustering algorithm. In particular, lexical stress was introduced as a clustering variable in addition to the phonetic context. The parameter sharing model was extended by tying HMM states across different target phones. For instance, one ...

This paper aims to present the newly designed acoustic emission AE instrument for the in-situ monitoring of corrosion damages of aircraft structures. The work was theoretically based on the combination of modal AE theory and AE parameter analysis technique. Previous research had shown that identification of corrosion AE and background noise by modal acoustic emission(MAE) technique was possible...