ZASPiL 40 = Speech production and perception : Experimental analyses and models
Refine
Year of publication
- 2005 (14)
Document Type
- Part of a Book (14)
Language
- English (14)
Has Fulltext
- yes (14)
Is part of the Bibliography
- no (14)
Keywords
- Artikulation (6)
- Artikulatorische Phonetik (5)
- Artikulator (4)
- Phonetik (3)
- Akustische Phonetik (2)
- Auditive Phonetik (2)
- Computerlinguistik (2)
- Experimentelle Phonetik (2)
- Kernspintomographie (2)
- Lautsprache (2)
It has been shown that visual cues play a crucial role in the perception of vowels and consonants. Conflicting consonantal stimuli presented in the visual and auditory modalities can even result in the emergence of a third perceptual unit (McGurk effect). From a developmental point of view, several studies report that newborns can associate the image of a face uttering a given vowel to the auditory signal corresponding to this vowel; visual cues are thus used by the newborns. Despite the large number of studies carried out with adult speakers and newborns, very little work has been conducted with preschool-aged children. This contribution is aimed at describing the use of auditory and visual cues by 4 and 5-year-old French Canadian speakers, compared to adult speakers, in the identification of voiced consonants. Audiovisual recordings of a French Canadian speaker uttering the sequences [aba], [ada], [aga], [ava], [ibi], [idi], [igi], [ivi] have been carried out. The acoustic and visual signals have been extracted and analysed so that conflicting and non-conflicting stimuli, between the two modalities, were obtained. The resulting stimuli were presented as a perceptual test to eight 4 and 5-year-old French Canadian speakers and ten adults in three conditions: visual-only, auditory-only, and audiovisual. Results show that, even though the visual cues have a significant effect on the identification of the stimuli for adults and children, children are less sensitive to visual cues in the audiovisual condition. Such results shed light on the role of multimodal perception in the emergence and the refinement of the phonological system in children.
Four speakers repeated 8 times 15 sentences containing 'pVp' syllables (V being /a/, /i/ and /u/). The 'pVp' syllables were located in final, penultimate and antepenultimate position relatively to the Intonational Phrase (IP) boundary. They were embedded in lexical words of 1-3 syllables and were either word-initial or word-final. Results show that the closer the vowel in word-final position is to the IP boundary, the longer the duration and the higher the fundamental frequency of the vowel; it is also characterised by larger lip opening gestures. The potential reduction or coarticulation of vowels in wordinitial position compared to their counterparts in word-final position is discussed.
In this paper the issue of the nature of the representations of the speech production task in the speaker's brain is addressed in a production-perception interaction framework. Since speech is produced to be perceived, it is hypothesized that its production is associated for the speaker with the generation of specific physical characteristics that are for the listeners the objects of speech perception. Hence, in the first part of the paper, four reference theories of speech perception are presented, in order to guide and to constrain the search for possible correlates of the speech production task in the physical space: the Acoustic Invariance Theory, the Adaptive Variability Theory, the Motor Theory and the Direct-Realist Theory. Possible interpretations of these theories in terms of representations of the speech production task are proposed and analyzed. In a second part, a few selected experimental studies are presented, which shed some light on this issue. In the conclusion, on the basis of the joint analysis of theoretical and experimental aspects presented in the paper, it is proposed that representations of the speech production task are multimodal, and that a hierarchy exists among the different modalities, the acoustic modality having the highest level of priority. It is also suggested that these representations are not associated with invariant characteristics, but with regions of the acoustic, orosensory and motor control spaces.
The contribution of von Kempelen's "Mechanism of Speech" to the 'phonetic sciences' will be analyzed with respect to his theoretical reasoning on speech and speech production on the one hand and on the other in connection with his practical insights during his struggle in constructing a speaking machine. Whereas in his theoretical considerations von Kempelen's view is focussed on the natural functioning of the speech organs – cf. his membraneous glottis model – in constructing his speaking machine he clearly orientates himself towards the auditory result – cf. the bag pipe model for the sound generator used for the speaking machine instead. Concerning vowel production his theoretical description remains questionable, but his practical insight that vowels and speech sounds in general are only perceived correctly in connection with their surrounding sounds – i.e. the discovery of coarticulation – is clearly a milestone in the development of the phonetic sciences: He therefore dispenses with the Kratzenstein tubes, although they might have been based on more thorough acoustic modelling.
Finally, von Kempelen's model of speech production will be discussed in relation to the discussion of the acoustic nature of vowels afterwards [Willis and Wheatstone as well as von Helmholtz and Hermann in the 19th century and Stumpf, Chiba & Kajiyama as well as Fant and Ungeheuer in the 20th century].
This paper presents the results of Open Quotient measurements in EGG signals of young (18 to 30 year old) and elderly (59 to 82 year old) male and female speakers. The paper further presents quantitative results on the relation between the OQ and the perception of a speaker's age. Higgins & Saxman (1991) found a decreased OQEGG with increasing age for females, whereas the OQEGG in sustained vowel material increased for males as the speakers age increased. In Linville (2002), however, the spectral amplitudes in the region of F0 (obtained by LTAS-measurements of read speech material) increased with increasing age independent of gender; this could be interpreted indirectly as an increasing OQ. We measured the OQEGG not only for sustained vowels, but also in vowels taken from isolated words. In order to analyse the relation between breathiness in terms of an increased OQ and the mean perceived age per stimulus a perception test was carried out in which listeners were asked to estimate speaker's age based on sustained /a/-vowel stimuli varying in vocal effort (soft - normal - loud) during production. The results indicated the following: (i) The decreased OQ for elderly females originally found by Higgins & Saxman is not apparent in our data for sustained /a/-vowels. For our female speakers no significant difference between the OQ of young and old speakers was found; for elderly males, however, we also found an increasing OQ with increasing age.(ii) In addition, a statistically significant increased OQEGG occurs for the group of the elderly males for the vowels from the word material. (iii) Our results show a strong positive relation between perceived age and OQ in male voices. Regarding (i) and (ii), at least the male speaker's voice becomes more breathy as age increases. Considering (iii), increased breathiness may contribute to the listener’s perception of increased age.
This paper describes the processing of MRI and CT images needed for developing a 3D linear articulatory model of velum. The 3D surface that defines each organ constitutive of the vocal and nasal tracts is extracted from MRI and CT images recorded on a subject uttering a corpus of artificially sustained French vowels and consonants. First, the 2D contours of the organs have been manually extracted from the corresponding images, expanded into 3D contours, and aligned in a common 3D coordinate system. Then, for each organ, a generic mesh has been chosen and fitted by elastic deformation to each of the 46 3D shapes of the corpus. This has finally resulted in a set of organ surfaces sampled with the same number of 3D vertices for each articulation, which is appropriate for Principal Component Analysis or linear decomposition. The analysis of these data has uncovered two main uncorrelated articulatory degrees of freedom for the velum's movement. The associated parameters are used to control the model. We have in particular investigated the question of a possible correlation between jaw / tongue and velum's movement and have not find more correlation than the one found in the corpus.
This paper contributes to the understanding of vocal folds oscillation during phonation. In order to test theoretical models of phonation, a new experimental set-up using a deformable vocal folds replica is presented. The replica is shown to be able to produce self sustained oscillations under controlled experimental conditions. Therefore different parameters, such as those related to elasticity, to acoustical coupling or to the subglottal pressure can be quantitatively studied. In this work we focused on the oscillation fundamental frequency and the upstream pressure in order to start (on-set threshold) either end (off-set threshold) oscillations in presence of a downstream acoustical resonator. As an example, it is shown how this data can be used in order to test the theoretical predictions of a simple one-mass model.
A visual articulatory model and its application to therapy
of speech disorders : a pilot study
(2005)
A visual articulatory model based on static MRI-data of isolated sounds and its application in therapy of speech disorders is described. The model is capable of generating video sequences of articulatory movements or still images of articulatory target positions within the midsagittal plane. On the basis of this model (1) a visual stimulation technique for the therapy of patients suffering from speech disorders and (2) a rating test for visual recognition of speech movements was developed. Results indicate that patients produce recognition rates above level of chance already without any training and that patients are capable of increasing their recognition rate over the time course of therapy significantly.
This paper summarizes our research efforts in functional modelling of the relationship between the acoustic properties of vowels and perceived vowel quality. Our model is trained on 164 short steady-state stimuli. We measured F1, F2, and additionally F0 since the effect of F0 on perceptual vowel height is evident. 40 phonetically skilled subjects judged vowel quality using the Cardinal Vowel diagram. The main focus is on refining the model and describing its transformation properties between the F1/F2 formant chart and the Cardinal Vowel diagram. An evaluation of the model based on 48 additional vowels showed the generalizability of the model and confirmed that it predicts perceived vowel quality with sufficient accuracy.
The goal of our current project is to build a system that can learn to imitate a version of a spoken utterance using an articulatory speech synthesiser. The approach is informed and inspired by knowledge of early infant speech development. Thus we expect our system to reproduce and exploit the utility of infant behaviours such as listening, vocal play, babbling and word imitation. We expect our system to develop a relationship between the sound-making capabilities of its vocal tract and the phonetic/phonological structure of imitated utterances. At the heart of our approach is the learning of an inverse model that relates acoustic and motor representations of speech. The acoustic to auditory mappings uses an auditory filter bank and a self-organizing phase of learning. The inverse model from auditory to vocal tract control parameters is estimated using a babbling phase, in which the vocal tract is essentially driven in a random manner, much like the babbling phase of speech acquisition in infants. The complete system can be used to imitate simple utterances through a direct mapping from sound to control parameters. Our initial results show that this procedure works well for sounds generated by its own voice. Further work is needed to build a phonological control level and achieve better performance with real speech.