Quantitative systematic analysis of vocal tract data

Silva, Samuel; Teixeira, António

doi:10.1016/j.csl.2015.05.004

Cited by 10 publications

(5 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With this set of variables we move away from choosing fixed points over the tongue, as in previous work. Instead, maximal constrictions are determined between different tract segments (e.g., tongue tip and hard palate, identified based on the segmentation data [16]) as follows:…”

Section: Choice and Computation Of Tract Variablesmentioning

confidence: 99%

Data-Driven Critical Tract Variable Determination for European Portuguese

et al. 2020

Self Cite

View full text Add to dashboard Cite

Technologies, such as real-time magnetic resonance (RT-MRI), can provide valuable information to evolve our understanding of the static and dynamic aspects of speech by contributing to the determination of which articulators are essential (critical) in producing specific sounds and how (gestures). While a visual analysis and comparison of imaging data or vocal tract profiles can already provide relevant findings, the sheer amount of available data demands and can strongly profit from unsupervised data-driven approaches. Recent work, in this regard, has asserted the possibility of determining critical articulators from RT-MRI data by considering a representation of vocal tract configurations based on landmarks placed on the tongue, lips, and velum, yielding meaningful results for European Portuguese (EP). Advancing this previous work to obtain a characterization of EP sounds grounded on Articulatory Phonology, important to explore critical gestures and advance, for example, articulatory speech synthesis, entails the consideration of a novel set of tract variables. To this end, this article explores critical variable determination considering a vocal tract representation aligned with Articulatory Phonology and the Task Dynamics framework. The overall results, obtained considering data for three EP speakers, show the applicability of this approach and are consistent with existing descriptions of EP sounds.

show abstract

Section: Choice and Computation Of Tract Variablesmentioning

confidence: 99%

Data-Driven Critical Tract Variable Determination for European Portuguese

et al. 2020

Self Cite

View full text Add to dashboard Cite

show abstract

“…The vocal tract outlines were extracted adopting the method proposed by Silva et al [17], resulting in contours identifying the different regions of interest, as depicted in Figure 1.…”

Section: Vocal Tract Data Processing and Analysismentioning

confidence: 99%

“…The comparison among vocal tract configurations was performed adopting and extending a previously proposed framework (for the sake of brevity, additional details can be found in [18,17,19]) enabling normalized quantification of differences, for different articulators/regions of the vocal tract and their visualization (in a diagram as illustrated in Fig. 1: velum (VEL), tongue dorsum (TD), tongue back (TB), tongue tip (TT), lip protrusion (LP), lip aperture (LA) and pharynx (Ph).…”

Section: Vocal Tract Data Processing and Analysismentioning

confidence: 99%

On the Role of Oral Configurations in European Portuguese Nasal Vowels

Cunha¹,

Silva²,

Teixeira³

et al. 2019

Interspeech 2019

Self Cite

View full text Add to dashboard Cite

The characterisation of nasal vowels is not only a question of studying velar aperture. Recent work shows that oropharyngeal articulatory adjustments enhance the acoustics of nasal coupling or, at least, magnify differences between oral/nasal vowel congeners. Despite preliminary studies on the oral configurations of nasal vowels, for European Portuguese, a quantitative analysis is missing, particularly one to be applied systematically to a desirably large number of speakers. The main objective of this study is to adapt and extend previous methodological advances for the analysis of MRI data to further investigate: how velar changes affect oral configurations; the changes to the articulators and constrictions when compared with oral counterparts; and the closest oral counterpart. High framerate RT-MRI images (50fps) are automatically processed to extract the vocal tract contours and the position/configuration for the different articulators. These data are processed by evolving a quantitative articulatory analysis framework, previously proposed by the authors, extended to include information regarding constrictions (degree and place) and nasal port. For this study, while the analysis of data for more speakers is ongoing, we considered a set of two EP native speakers and addressed the study of oral and nasal vowels mainly in the context of stop consonants.

show abstract

“…A variety of MR sequences are now in use for vocal tract imaging, with each providing different extents of spatial and temporal resolution, as well as varying artifact sensitivity. Commonly used 2D acquisitions include Cartesian sequences (single-slice gradient echo, e.g., Silva & Teixeira, 2015a) spiral acquisitions (e.g., Bresch & Narayanan, 2009;Narayanan et al, 2004), and radial acquisitions (e.g., Niebergall et al, 2013). Approaches to imaging the vocal tract in 3D have typically employed multi-slice T1w acquisitions (e.g., Badin et al, 2002;Baer et al, 1991).…”

Section: Vocal Tract Mr -Sequences and Acquisition Parametersmentioning

confidence: 99%

“…More elaborate computational approaches have been developed recently, imposing specific model predictions concerning vocal tract morphology and articulator position (e.g., Bresch & Narayanan, 2009;Silva & Teixeira, 2015a). In one such approach, the model is trained a priori on a corpus of real-time images; vocal tract data from a given subject may then be segmented based on the model's assumptions regarding vocal tract morphology (Silva & Teixeira, 2015b).…”

Section: Vocal Tract Mr -Automated Segmentation and Analysesmentioning

confidence: 99%

Magnetic resonance imaging of the brain and vocal tract: Applications to the study of speech production and language learning

Carey

McGettigan

2017

Neuropsychologia

View full text Add to dashboard Cite

The human vocal system is highly plastic, allowing for the flexible expression of language, mood and intentions. However, this plasticity is not stable throughout the life span, and it is well documented that adult learners encounter greater difficulty than children in acquiring the sounds of foreign languages. Researchers have used magnetic resonance imaging (MRI) to interrogate the neural substrates of vocal imitation and learning, and the correlates of individual differences in phonetic "talent". In parallel, a growing body of work using MR technology to directly image the vocal tract in real time during speech has offered primarily descriptive accounts of phonetic variation within and across languages. In this paper, we review the contribution of neural MRI to our understanding of vocal learning, and give an overview of vocal tract imaging and its potential to inform the field. We propose methods by which our understanding of speech production and learning could be advanced through the combined measurement of articulation and brain activity using MRI - specifically, we describe a novel paradigm, developed in our laboratory, that uses both MRI techniques to for the first time map directly between neural, articulatory and acoustic data in the investigation of vocalisation. This non-invasive, multimodal imaging method could be used to track central and peripheral correlates of spoken language learning, and speech recovery in clinical settings, as well as provide insights into potential sites for targeted neural interventions.

show abstract

Quantitative systematic analysis of vocal tract data

Cited by 10 publications

References 23 publications

Data-Driven Critical Tract Variable Determination for European Portuguese

Data-Driven Critical Tract Variable Determination for European Portuguese

On the Role of Oral Configurations in European Portuguese Nasal Vowels

Magnetic resonance imaging of the brain and vocal tract: Applications to the study of speech production and language learning

Contact Info

Product

Resources

About