Zapewniamy

kompleksową diagnostykę wymowy
redukcję polskiego akcentu
poznanie dźwięków języka angielskiego
specjalistyczny trening wymowy
trening estetycznej artykulacji

Do pierwszego roku życia niemowlaki są w stanie rozróżnić wszystkie dźwięki mowy występujące w językach.dalej
Brytyjski angielski ma 11 samogłosek (bez samogłoski redukowanej w nieakcentowanych sylabach). Polski ma 6 samogłosek (bez unosowienia). Co więcej, nie ma w polskim samogłoski, którą byśmy mogli bezpośrednio przenieść do wymowy angielskiej. Wszystkie angielskie samogłoski są inne niż ...dalej
Bogactwo dźwięków mowy w językach świata jest niewiarygodne. Klasyfikuje się 558 spółgłosek, 260 samogłosek oraz 51 dyftongówdalej

PolGem

wielkość tekstu: A | A | A

POLGEM - the recorded corpus of Polish geminate consonants

Arkadiusz Rojczyk
Andrzej Porzuczek
Speech Processing Laboratory, University of Silesia in Katowice
(Acknowledged assistance by Monika Piwowarska)

The project supported by the National Science Centre Poland grant Acoustic properties of Polish geminate consonants 2017/25/B/HS2/02548 (Principal Investigator: Arkadiusz Rojczyk)

Please cite: Rojczyk, A., Porzuczek, A. (2022). POLGEM - The recorded corpus of Polish geminate consonants. Poznan Studies in Contemporary Linguistics 58(2): 253-262, https://doi.org/10.1515/psicl-2022-0014

1. Introduction

POLGEM is a recorded corpus of Polish orthographic geminates that we make available to all researchers investigating geminates in the world's languages. It contains the recorded productions of 111 Polish words by 54 native speakers of Polish (5994 tokens altogether). The recordings are segmented into individual words in wav files. They may be used for both acoustic analyses and perception experiments.

2. Structure

The Excel file below contains detailed information on the recorded words. They are annotated using the following categories:
Word
Transcription
English translation
Consonant
Consonant class
Voicing
Word position
Lexical/Morphological
Borrowing YES/NO
Borrowing (language)

1	2	3	4	5	6	7	8	9	10
Word	Transcription	English translation	Consonant	Consonant class	Voicing	Word position	Lexical/ Morphological	Borrowing YES/NO	Borrowing (language)[1]
Abba	'abba	Abba	b	plo	+	MED	L	Y	SWE
Addis Abeba	a(d)dis a'bɛba	Addis Abeba	d	plo	+	MED	L	Y	AMH
Aleppo	a'lɛppɔ	Aleppo	p	plo	-	MED	L	Y	ITA[2]
allegro	al'lɛgrɔ	allegro	l	app	+	MED	L	Y	ITA
alleluja	allɛ'luja	hallellujah	l	app	+	MED	L	Y	HEB
annały	an'nawɨ	annals	n	nas	+	MED	L	Y	LAT
Annasz	'annaʂ	Annas	n	nas	+	MED	L	Y	HEB
attyka	at'tɨka	attic	t	plo	-	MED	L	Y	GRE
ballada	bal'lada	ballad	l	app	+	MED	L	Y	FRA
bessa	'bɛssa	slump	s	fri	-	MED	L	Y	FRA
bezzębny	bɛz'zɛmbnɨ	toothless	z	fri	+	MED	M	N
biennale	bjɛn'nalɛ	biennial	n	nas	+	MED	L	Y	ITA
Budda	'budda	Buddha	d	plo	+	MED	L	Y	SKR
bulla	'bulla	(pope's) bull	l	app	+	MED	L	Y	LAT
confetti	kɔn'fɛtti	confetti	t	plo	-	MED	L	Y	ITA
corrida	kɔr'rida	bullfight	r	tap	+	MED	L	Y	ESP
czczo	tʂtʂɔ	empty (ADV)	tʂ	aff	-	INI	L	N
czczy	tʂtʂɨ	empty	tʂ	aff	-	INI	L	N
czynny	'tʂɨnnɨ	active	n	nas	+	MED	M	N
denny	'dɛnnɨ	demersal	n	nas	+	MED	M	N
dziecinny	dʑɛ'tɕinnɨ	childish	n	nas	+	MED	M	N
dziennik	'dʑɛɲɲik	diary	n	nas	+	MED	M	N
dziewanna	dʑɛ'vanna	mullein	n	nas	+	MED	L	N
dżdżownica	dʐdʐɔv'nitsa	earthworm	dʐ	aff	+	INI	L	N
dżdżysty	'dʐdʐɨstɨ	drizzly	dʐ	aff	+	INI	L	N
errata	ɛr'rata	errata	r	tap	+	MED	L	Y	LAT
falliczny	fal'litʂnɨ	phallic	l	app	+	MED	L	Y	GRE
flotylla	flɔ'tɨlla	flotilla	l	app	+	MED	L	Y	ESP
gamma	'gamma	gamma	m	nas	+	MED	L	Y	GRE
gehenna	gɛ'hɛnna	Gehenna	n	nas	+	MED	L	Y	HEB
getto	'gɛttɔ	ghetto	t	plo	-	MED	L	Y	ITA
glissando	gli(s)'sandɔ	glissando	s	fri	-	MED	L	Y	FRA
gminny	'gminnɨ	borough (ADJ)	n	nas	+	MED	M	Y	GER
greccy	'rɛtstsɨ	Greek PL	ts	aff	-	MED	M	N
helleński	hɛl'lɛɲski	Hellenic	l	app	+	MED	L	Y	GRE
hi(p)pis	'hi(p)pis	hippy	p	plo	-	MED	L	Y	ENG
hippika	hip'pika	equestrianism	p	plo	-	MED	L	Y	GRE
hobby	'hɔbbɨ	hobby	b	plo	+	MED	L	Y	ENG
horror	'hɔrrɔr	horror	r	tap	+	MED	L	Y	ENG
hossa	'hɔssa	boom	s	fri	-	MED	L	Y	FRA
hostessa	hɔs'tɛssa	hostess	s	fri	-	MED	L	Y	FRA
idylla	i'dɨlla	idyll	l	app	+	MED	L	Y	GRE
immanentny	imma'nɛntnɨ	immanent	m	nas	+	MED	M	Y	LAT
immatrykulacja	immatrɨku'latsja	matriculation	m	nas	+	MED	M	Y	LAT
immunitet	immu'ɲitɛt	immunity	m	nas	+	MED	M	Y	LAT
immunologia	immunɔ'lɔgja	immunology	m	nas	+	MED	M	Y	LAT/GRE
inny	'innɨ	other	n	nas	+	MED	M	N
Ja(f)fa	'ja(f)fa	Jaffa	f	fri	-	MED	L	Y	HEB
jedźcie	'jɛtɕtɕɛ	ride (IMP, PL)	tɕ	aff	-	MED	M	N
konny	'kɔnnɨ	equestrian	n	nas	+	MED	M	N
lasso	'lassɔ	lasso	s	fri	-	MED	L	Y	ESP
Lavazza	la'vatstsa	Lavazza	ts	aff	-	MED	L	Y	ITA
lekki	'lɛkki	light	k	plo	-	MED	L	N
lenno	'lɛnnɔ	fief	n	nas	+	MED	L	Y	GER
lobby	'lɔbbɨ	lobby	b	plo	+	MED	L	Y	ENG
lotto	'lɔttɔ	lotto	t	plo	-	MED	L	Y	ITA
Lozanna	lɔ'zanna	Lausanne	n	nas	+	MED	L	Y	FRA
madonna	ma'dɔnna	madonna	n	nas	+	MED	L	Y	ITA
maggi	'magi	stock cube	g	plo	+	MED	L	Y	ITA
mammografia	mammɔ'grafja	mammography	m	nas	+	MED	L	Y	LAT/GRE
manna	'manna	manna, semolina	n	nas	+	MED	L	Y	GRE
Marzanna	ma'ʐanna	a female name	n	nas	+	MED	L	N
Marzenna	ma'ʐɛnna	a female name	n	nas	+	MED	L	N
Mekka	'mɛkka	Mecca	k	plo	-	MED	L	Y	ARB
mełła	'mɛwwa	(She) milled	w	app	+	MED		N
mennica	mɛn'ɲitsa	mint (a place)	n	nas	+	MED	L	Y	GER
messa	'mɛssa	mess	s	fri	-	MED	L	Y	GER
miękki	'mjɛŋ(k)ki	soft	k	plo	-	MED	L	N
mułła	'muwwa	mullah	w	app	+	MED	L	Y	TUR
najjaśniejszy	najjaɕ'ɲɛjʂɨ	the brightest	j	app	+	MED	M	N
narracja	nar'ratsja	narration	r	tap	+	MED	L	Y	LAT
niewinnie	ɲɛ'viɲɲɛ	innocent (ADV)	ɲ	nas	+	MED	M	N
obronna	ɔ'brɔnna	defensive	n	nas	+	MED	M	N
oddać	'ɔddatɕ	give back	d	plo	+	MED	M	N
oddech	'ɔddɛx	breath	d	plo	+	MED	M	N
panna	'panna	maiden	n	nas	+	MED	M	N
passa	'passa	streak	s	fri	-	MED	L	Y	FRA
pełła	'pɛwwa	(She) weeded	w	app	+	MED		N
piccolo	'pi(k)kɔlɔ	piccolo	k	plo	-	MED	L	Y	ITA
pizza	'pitstsa	pizza	ts	aff	-	MED	L	Y	ITA
pizzeria	pits'tsɛrja	pizzeria	ts	aff	-	MED	L	Y	ITA
płonna	'pwɔnna	vain	n	nas	+	MED	M	N
płynny	'pwɨnnɨ	liquid	n	nas	+	MED	M	N
poddać	'pɔddatɕ	submit	d	plo	+	MED	M	N
Pyrrus	'pɨrrus	Pyrrhus	r	tap	+	MED	L	Y	GRE
rattan	'ra(t)tan	rattan	t	plo	-	MED	L	Y	ENG
rozzuć	'rɔzzutɕ	take off (sb's shoes)	z	fri	+	MED	M	N
salmonella	salmɔ'nɛlla	salmonella	l	app	+	MED	L	Y	ENG
sawanna	sa'vanna	savanna	n	nas	+	MED	L	Y	ESP
senna	'sɛnna	sleepy (FEM)	n	nas	+	MED	M	N
Sewilla	sɛ'villa	Seville	l	app	+	MED	L	Y	ESP
skłonny	'skwɔnnɨ	prone	n	nas	+	MED	M	N
spaghetti	spa'gɛtti	spaghetti	t	plo	-	MED	L	Y	ITA
ssać	ssatɕ	suck	s	fri	-	INI	L	N
ssak	ssak	mammal	s	fri	-	INI	L	N
surrealizm	surrɛ'alizm	surrealism	r	tap	+	MED	M	Y	FRA
sutanna	su'tanna	cassock	n	nas	+	MED	L	Y	ITA
świeccy	'ɕfjɛtstsɨ	laymen	ts	aff	-	MED	M	N
terror	'tɛrrɔr	terror	r	tap	+	MED	L	Y	LAT
to(r)reador	tɔ(r)rɛ'adɔr	toreador	r	tap	+	MED	L	Y	ESP
Toro Rosso	tɔrɔ 'rɔssɔ	Toro Rosso/Red Bull	s	fri	-	MED	L	Y	ITA
tutti-frutti	tutti 'frutti	tutti-frutti	t	plo	-	MED	L	Y	ITA
uczczą	'utʂtʂõ	(They) will celebrate	tʂ	aff	-	MED		N
vendetta	vɛn'dɛtta	vendetta	t	plo	-	MED	L	Y
wanna	'vanna	bathtub	n	nas	+	MED	M	Y	GER
willa	'villa	villa	l	app	+	MED	L	Y	LAT
winnica	viɲ'ɲitsa	vineyard	ɲ	nas	+	MED	M	N
wwozić	'vvɔʑitɕ	bring in	v	fri	+	INI	M	N
zmienna	'zmjɛnna	variable	n	nas	+	MED	M	N
zzuć	zzutɕ	take off (shoes)	z	fri	+	INI	M	N

[1] We used Tokarski (1980) to verify word origins.

[2] Where possible, we indicate the direct loan language rather than the word origin, which, especially in the case of geographical names, may be disputable.

Structure Excel file

3. Speakers

The following Excel file contains speaker data with gender and age:

Speaker Excel file

4. Analysis and coding

All recordings were analysed to determine three aspects of each produced token:

Degemination – in contemporary Polish some words undergo degemination, which may emerge in some specific words or be speaker-specific. In order to facilitate future work with our database, the degeminated tokens were coded in the file name with an appended _S label. Of the total of 5994 recorded words, 638 (11%) were degeminated and realised with a singleton. The most frequently degeminated words were piccolo ‘piccolo’ (83%), hippis ‘hippi’, (76%), maggi ‘magi’ (74%), glissando ‘glissando’ (61%). All these words are borrowings from Italian, English and French. All speakers produced at least one degeminated token, however some speakers were more prone to degeminate than others. For example, Speaker 9 produced only one degeminated word while Speaker 53 produced as many as 34 degeminated words.
Rearticulation – all recorded words were analysed to detect rearticulation in geminate consonants. Rearticulation occurred in 1231 words, yielding a proportion of 20% in the total of 5994 words. Similar to the report by Rojczyk and Porzuczek (2019), there is an observable between-speaker variability in the rearticulation rate with the range of 3% (Speaker 28) to 54% (Speaker 52).
Mispronunciation – words with audible mispronunciations were coded by appending _MIS to the file name of each mispronounced word. The number of mispronounced productions was very low, only 15 words were mispronounced, yielding the proportion of 0.2%.

Analysis Excel file

4. Recordings

Speaker 1
Speaker 2
Speaker 3
Speaker 4
Speaker 5
Speaker 6
Speaker 7
Speaker 8
Speaker 9
Speaker 10
Speaker 11
Speaker 12
Speaker 13
Speaker 14
Speaker 15
Speaker 16
Speaker 17
Speaker 18
Speaker 19
Speaker 20
Speaker 21
Speaker 22
Speaker 23
Speaker 24
Speaker 25
Speaker 26
Speaker 27
Speaker 28
Speaker 29
Speaker 30
Speaker 31
Speaker 32
Speaker 33
Speaker 34
Speaker 35
Speaker 36
Speaker 37
Speaker 38
Speaker 39
Speaker 40
Speaker 41
Speaker 42
Speaker 43
Speaker 44
Speaker 45
Speaker 46
Speaker 47
Speaker 48
Speaker 49
Speaker 50
Speaker 51
Speaker 52
Speaker 53
Speaker 54

References:

Rojczyk, A. and A. Porzuczek. (2019). “Durational properties of Polish geminate consonants”. Journal of the Acoustical Society of America 146(6). 4171-4182.

Tokarski, J. (ed.). 1980. Słownik wyrazów obcych PWN. Warszawa: Państwowe Wydawnictwo Naukowe.

Badania

PolGem