Articulatory patterns of Russian diphthongised
vowels: pilot MRI investigation
G. Kedrova
1 Lomonosov
1,
N. Anisimov
1,
V. Ushakov2
Moscow State University; 2NRC Kurchatov Institute
Russian Federation
ICPhS 2015 / P2.37
The research is aimed at the experimental MRI-investigation of
model articulatory gestures corresponding to articulatory
patterns of the Russian vowel phoneme /ɨ/ ([ы]) .
1. Introduction
At present, there is general discordance of opinions discussing articulatory patterns of diphthongised /
non-diphthongised Russian vowel /ɨ/ ([ы]). Main generally accepted hypotheses on articulatory
mechanism of the vowel’s /ɨ/ ([ы]) production are as follows.
Hypothesis 1: vowel /ɨ/ ([ы]) in comparison to vowel /i/ ([и])
According to R. Avanesov: “…in both cases tongue dorsum is raised high towards palate, but for /i/ an
overall tongue movement is realized with its medial part, whereas for /ɨ/ it’s the tongue back that is
more active. Tongue tip during pronunciation of /i/ is close to lower teeth, whereas for /ɨ/ the whole
tongue body is retracted backwards and the tongue tip is slightly raised. Switching from pronunciation
of /i/ to pronunciation of /ɨ/ one could easily feel displacement of the focus of articulation backward into
interior part of her/his oral cavity.” [1].
Recent X-ray data of the vowel’s articulatory contours tend to support Avanesov’s description (though
also without indication of the sound’s articulatory stability / instability) [2].
X-ray data of
L.Skalozub (nondiphthongised
.
vowel /ɨ/ [ы]).
X-ray data of
L.Skalozub (vowel
/i/ [и]).
Among our vowels, pick out the YI, born in the Mongol bowels...
Joseph Brodsky. “Portrait of Tragedy”
Из гласных, идущих горлом, выбери Ы, придуманное монголом…
Иосиф Бродский. «Портрет трагедии»
2. Material and Methods
MRI experiments #1 were realized on a 0.5 T MR scanner (Tomikon S50 “Bruker”). MR scanning was
executed on sagittal plane with the slice thickness of 9 mm and to a field of view 200*120 mm. The pulse
sequence 'gradient echo' was used with the following parameters: TR=12 ms, TE=5.5 ms, FA=10
degrees.
Result: 2.7 frames in a second and with 3 mm in-plane resolution.
MRI experiments #2 were realized on a 3 T MR scanner (Magnetom Verio 3T “Siemens”) with following
parameters: MR scanning was executed on sagittal plane with the slice thickness of 10 mm and to a field
of view 130*130 mm. The pulse sequence ‘gradient echo’ was used with the following parameters:
TR=3.5 ms, TE=1.48 ms, FA=10 degrees.
Result: 4-5 MR images per second.
Irrespective of the MRI sessions extra control audio and video recordings of the same speech data from
the same reference subject producing speech in the same position were realized in a record's studio
environment. The whole data set of MR images collected in all experimental acquisitions was identified
and ascribed to each phase of each phoneme realization.
Hypothesis 2: context dependent variants of the vowel /ɨ/ ([ы])
Matusevič (upon L. Ščerba) [3] describes two context dependent modes of articulation of the vowel /ɨ/
(context variant of non-diphthongised or the one as gliding sound).
close middle vowel
followed by nonpalatalized
consonant as in /sɨn/
(son), /bɨk/ (bull),
/mɨ/ (we),
gliding (diphthongized)
middle vowel followed
by palatalized
consonant as in /bɨt’/
(to be), /mɨt’/ (to wash).
Hypothesis 3: vowels /ɨ/ ([ы]) and /u/ ([у]) as diphthongised Russian vowels
The Russian phonological system has no diphthongs as linguistic phonetic units, however most
researchers acknowledge gliding articulatory gestures during pronunciation of some Russian vowels,
first and foremost close central unrounded vowel /ɨ/ ([ы]), rounded back vowel /o/ ([о]) [4] and close
back vowel /u/ ([у]) [5] .
X-ray data of
K.Bolla’s “Atlas of
the Russian speech
sounds”
(diphthongised
vowel /ɨ/)
Experimental data of MRI acquisitions’ interface, post-MRI session control audio and
video recordings incorporated. [6]
X-ray data of
K.Bolla’s “Atlas of
the Russian speech
sounds”
(diphthongised
vowel /u/)
4. Conclusions
3. Results
In order to make more evident comparative appraisal of the vowel /ɨ/’s articulation dynamics each typical
sagittal cut slice has been ascribed appropriate contour of the vocal tract’s relevant active and passive
organs’ configurations. These contours were manually traced upon corresponding MR images and
superimposed within the same frame, the overlapping being aligned upon several fixed anatomical
landmarks (upon the individual geometry of the speaker’s maxilla as well as configuration of his/her
vertebral column). Sample MR images of sagittal articulatory contours for vowel’s /ɨ/ ([ы]) production
phases in pronunciation of three experimental subjects are exposed below.
speaker #1
speaker #2
speaker #3
I. The current research’s MRI data revealed two distinctly different dynamic patterns of the vowel’s /ɨ/
articulatory heterogeneity. The first one conforms to traditional descriptions of the vowel’s /ɨ/ articulation
as it has been previously described in manuals of Russian Phonetics [3, 4, 5]. This articulatory pattern is
implemented by drifting displacement of the tongue dorsum and tongue blade from central position
towards anterior part of the buccal cavity, gradually enlarging thus resonant airspace in the pharyngeal
area. Other experimental subjects, however, manifested very particular articulatory manner, different
from the first type though similar among them. Articulatory heterogeneity of the phoneme /ɨ/ in
pronunciation of speakers #1 and #2 is spatially and temporary determined by complex interaction of
several speech articulators. The interaction involves both tongue transformations (first and foremost of
tongue body and tongue blade’s displacement) and apparent peculiar activity of the speakers’ mandible.
Special palatal constriction’s dynamics characteristic of the phoneme’s /ɨ/ quality reflected in typical
configurations of gradually changing F2 and F3 tracks in acoustic signal is therefore adjusted first and
foremost through specific lower jaw’s displacement.
II. Some previously conducted research of the articulatory base of various languages also proved
substantial dynamic syncretism of lower jaw and tongue body movements (jaw-tongue coupling). It might
be detected through correlation between phonemes’ duration and extent of jaw’s articulatory activity: the
longer the sound the more the jaw is able to move [7]. This statement correlates well with experimental
data on inherent phoneme’s temporal patterns of the Russian vowels, vowel /ɨ/ in particular [8].
III. We hypothesise that the two types of unstable articulatory behaviours revealed in current study
correlate with two main types of articulatory patterns of palatalization in Russian: massive involvement of
the tongue body in articulation processes coupled with substantial raising of the tongue dorsum in the front
region of the oral cavity and restriction of the tongue root VS tongue blade activity that is realised in its
convex shape in alveolar and post-alveolar regions [6].)
IV. We suggest that exceptional position of the phoneme /ɨ/ within the Russian phonetic and phonological
system presented in ongoing discussions and debates whether /ɨ/ is a single vowel phoneme OR an
allophone of the Russian phoneme /i/ [9]) might be associated with peculiarities of its exceptional
articulatory pattern.
5. References
[1] Avanesov, R.I. 1972. Russkoje literaturnoye proiznosheniye [Russian standard pronunciation]. Moscow: Prosveshcheniye.
[2] Skalozub, L.G. 1979. Dinamika zvukoobrazovaniya (po dannym kinorentgenografirovaniya) [Dynamics of speech production
(upon X-ray data]. Kiev: Naukova dumka.
[3] Matusevič, M.N. 1976. Sovremenny russkiy yazyk. Fonetika [Modern Russian language: Phonetics]. Moscow: Prosveshcheniye.
[4] Bondarko, L.V. 1977. Zvukovoy stroi sovremennogo russkogo jazyka [Phonetic system of the modern Russian language].
Moscow: Prosveshcheniye.
[5] Bolla, K. 1981. Atlas zvukov russkoy rechi [Atlas of the Russian speech sounds]. Budapest.
[6] Kedrova, G., Anisimov, N., Zaharov, L. 2008. Magnetic Resonance investigation of palatalized stop consonants and spirants in
Russian, Proc. 7th European Conference on Noise Control 2008 Paris, 2345-2350.
[7] Mooshammer, C.; Hoole, P. & Geumann, A. 2007. Jaw and order. Language and Speech, 50, 145-176.
E-mail: [email protected] (G. Kedrova); [email protected] (N. Anisimov); Address: Russian Federation,
119991 Moscow, GSP-2, Leninskije Gory, MGU, 1st Humanities, R. 983
[8] Ščerba, L.V. 1983. Teoria russkogo pis’ma [Theory of the Russian writing]. Leningrad: Nauka.
[9] Zinder L.R. 1964. Vlijanije tempa rechi na obrazovanije otdel’nyh zvukov [Impact of the speech tempo on speech sounds
production]. Voprosy fonetiki. No. 325, 3-28.