Alain Marchal

Speech production and speech modelling

Section 1: Physiological Framework for the Speech Production Process.- Organization of the Articulatory System: Peripheral Mechanisms and Central Coordination.- Respiratory Activity in Speech.- Acquisition of Speech Production: the Achievement of Segmental Independence.- Section 2: Coarticuiation and Other Connected Speech Processes.- Segmental Reduction in Connected Speech in German: Phonological Facts and Phonetic Explanations.- V-C-V Lingual Coarticuiation and its Spatiotemporal Domain.- Section 3: Models of Articulatory-Acoustic Relationships.- Compensatory Articulation During Speech Evidence from the Analysis and Synthesis of Vocal-tract Shapes Using an Articulatory Model.- Articulatory Synthesis.- Articulatory-Acoustic Relationships in Fricative Consonants.- Articulatory-Acoustic-Phonetic Relations and Modelling, Regions and Modes.- Evidence for Nonlinear Sound Production Mechanisms in the Vocal Tract.- Section 4: Theories and Models of Articulatory Organization and Timing.- Testing Theories of Speech Production: Implications of Some Detailed Analyses of Variable Articulatory Data.- Speech as Audible Gestures.- Articulatory Perspectives of Speech Organization.- Speech Motor Timing.- The Acoustic and Physiologic Characteristics of Neurologically Impaired Speech Movements.- Explaining Phonetic Variation: A Sketch of the H and H Theory.

Accor: Instrumentation and Database for the Cross-Language Study of Coarticulation

An unavoidable problem in speech technology, particularly in the development of robust automatic speech recognition systems, is the extreme variability in the acoustic attributes of segments. Segments are highly sensitive to context and bear little resemblance to their intrinsic characteristics manifested when they are uttered in isolation. However, the problem can become tractable if we model the linguistic and physiological aspects of coarticulatory processes, the main source of systematic variability at the segmental level.

Regenerating the spectral shapes of [s] and [∫] from a limited set of articulatory parameters

This work was aimed at exploring articulatory‐acoustic relationships in the production of French fricatives. More precisely, an attempt was made to find out whether the spectral shapes of [s] and [■] can be regenerated from the x and y coordinates of three electromagnetic transducers affixed to the tongue in the midsagittal plane. The corpus was composed of the two fricatives /s/ and /■/ combined with the vowels /a/ and /i/ in sequences of the type /Vs■V/ and /V■sV/, and was read by one male native speaker of French. The spectrum regeneration was based on a statistical procedure which consisted of estimating the factors explaining the main part of the acoustic variance from the position of the transducers, by means of multiple linear regression. The articulatory‐acoustic correlations were high and allowed us to regenerate the fricative spectra with a good accuracy. The way in which the acoustic parameters varied as a function of the articulatory ones in the statistical model was in good agreement with data reported in previous works. The results support the idea that the tongue has relatively few degrees of freedom in the production of [s] and [■].

Assessment of an electromagnetometric system for the investigation of articulatory movements in speech production

Electromagnetometry can be used to track articulatory movements during speech production by means of small electromagnetic receivers attached to the articulators. This study was aimed at determining the measurement accuracy of the Movetrack system marketed by Botronic (Sweden). The average difference between a position transduced by Movetrack and the actual position, obtained from a precision mechanical device, was less than 0.4 mm, when the receiver was placed in the midsagittal plane and parallel to the transmitters at a distance of about 25 cm from each transmitter. Higher deviations were observed when the receiver was tilted with respect to the transmitters. Thus the device appears to be capable of delivering accurate measures of position for rigid anatomical structures like the jaw, which are not likely to produce tilt.

Cross‐language EPG data on lingual asymmetry

Conventional x‐ray views of the vocal tract give only limited information about the lingua‐palatal contacts occurring during speech production. From these types of data, it is not really possible to evaluate the real size and the exact shape of the vocal tract. When needed, the area at different sections must, in fact, be reconstructed using a mathematical model [S. Maeda, J. Acoust. Soc. Am. Suppl. 1 65, S22 (1979)]. A review of literature indicates that little attention is paid to the potential asymmetry of the articulatory gestures. This phenomenon appears, however, to be particularly crucial in speech pathology [Suzuki, Jpn. J. Oral Maxillo. Fac. Surg. 30, 45–54 (1984)], where lingual and palatal asymmetry is often very severe and causes a poor speech intelligibility. But, even for normal speech, there is some evidence that lingual asymmetry may occur [Hamlet et al., J. Acoust. Soc. Am. 79, 1164–1169 (1986)]. Moreover, it has also been suggested that an asymmetrical tongue placement would be necessary...

A database of subaquatic and hyperbaric speech : PSH/DISPE

The present article describes a French and English divers’ speech database, now implemented and available on a CDROM. This database complies with the SAM EUROPEC standards. It consists of acoustic signal files and associated files which provide information about the speakers and recording conditions. A database management system is used for an easy and fast retrieval of the speech items recorded at various depths and with specific diving equipment.