Ctrl+K
Ctrl+K

Polish CV dictionary v2.0.0#

  • Maintainer: Vox Communis

  • Language: Polish

  • Dialect: N/A

  • Phone set: Epitran

  • Number of words: 53,067

  • Phones: a b d d͡z d͡ʑ f i j k kʲ l m n p q r s t t͡s t͡ɕ u v w x z ŋ ɔ ɔ̃ ɕ ɛ ɛ̃ ɡ ɡʲ ɨ ɲ ʂ ʐ ʑ

  • License: CC-0

  • Compatible MFA version: v2.0.0

  • Citation:

@misc{Ahn_Chodroff_2022,
	author={Ahn, Emily and Chodroff, Eleanor},
	title={VoxCommunis Corpus},
	address={\url{https://osf.io/t957v}},
	publisher={OSF},
	year={2022},
	month={Jan}
}

Acoustic models

Installation#

Install from the MFA command line:

mfa model download dictionary polish_cv

Or download from the release page.

Intended use#

This dictionary is intended for forced alignment of Polish transcripts.

This dictionary uses the Epitran phone set for Polish, and was used in training the Polish Epitran acoustic model. Pronunciations can be added on top of the dictionary, as long as no additional phones are introduced.

Performance Factors#

When trying to get better alignment accuracy, adding pronunciations is generally helpful, especially for different styles and dialects. The most impactful improvements will generally be seen when adding reduced variants that involve deleting segments/syllables common in spontaneous speech. Alignment must include all phones specified in the pronunciation of a word, and each phone has a minimum duration (by default 10ms). If a speaker pronounces a multisyllabic word with just a single syllable, it can be hard for MFA to fit all the segments in, so it will lead to alignment errors on adjacent words as well.

Ethical considerations#

Deploying any Speech-to-Text model into any production setting has ethical implications. You should consider these implications before use.

Demographic Bias#

You should assume every machine learning model has demographic bias unless proven otherwise. For pronunciation dictionaries, it is often the case that transcription accuracy and lexicon coverage for the prestige variety modeled in this dictionary compared to other variants. If you are using this dictionary in production, you should acknowledge this as a potential issue.

IPA Charts#

Consonants#

Obstruent symbols to the left of are unvoiced and those to the right are voiced.

Manner

Labial

Labiodental

Alveolar

Retroflex

Palatal

Velar

Uvular

Nasal
Occurrences:
14,014
Examples:
* kim,:
[kʲ i m]
* muszą:
[m u ʂ ɔ̃]
* mysz,:
[m ɨ ʂ]
* myśli:
[m ɨ ɕ l i]
Occurrences:
17,936
Examples:
* pewny:
[p ɛ v n ɨ]
* rana,:
[r a n a]
* celny:
[t͡s ɛ l n ɨ]
* oknem:
[ɔ k n ɛ m]
Occurrences:
9,277
Examples:
* nił:
[ɲ i w]
* pięć:
[p j ɛ ɲ t͡ɕ]
* niemu:
[ɲ ɛ m u]
* broni:
[b r ɔ ɲ i]
Occurrences:
650
Examples:
* rękę,:
[r ɛ ŋ k ɛ̃]
* lękam:
[l ɛ ŋ k a m]
* lęku.:
[l ɛ ŋ k u]
* rękę:
[r ɛ ŋ k ɛ]

Stop
Occurrences:
14,498
Examples:
* pewny:
[p ɛ v n ɨ]
* płód.:
[p w u d]
* pięć:
[p j ɛ ɲ t͡ɕ]
* pomyś:
[p ɔ m ɨ ɕ]
Occurrences:
5,714
Examples:
* brał:
[b r a w]
* ryby.:
[r ɨ b ɨ]
* luba;:
[l u b a]
* bluzę:
[b l u z ɛ]
Occurrences:
19,146
Examples:
* też.:
[t ɛ ʐ]
* stosu:
[s t ɔ s u]
* twój!:
[t f u j]
* czaił:
[t ʂ a j w]
Occurrences:
8,627
Examples:
* ud.:
[u d]
* płód.:
[p w u d]
* drzwi:
[d ʐ v i]
* "dym:
[d ɨ m]
Occurrences:
10,737
Examples:
* król:
[k r u l]
* kasę:
[k a s ɛ]
* "jaką:
[j a k ɔ̃]
* koszt:
[k ɔ ʂ t]
Occurrences:
5,417
Examples:
* mego!:
[m ɛ ɡ ɔ]
* gumę:
[ɡ u m ɛ]
* godz.:
[ɡ ɔ d͡z]
* mózgu:
[m u z ɡ u]
Occurrences:
2
Examples:

Affricate
Occurrences:
5,110
Examples:
* pracę:
[p r a t͡s ɛ]
* celny:
[t͡s ɛ l n ɨ]
* ojcu.:
[ɔ j t͡s u]
* cen.:
[t͡s ɛ n]
Occurrences:
783
Examples:
* godz.:
[ɡ ɔ d͡z]
* żądzy:
[ʐ ɔ n d͡z ɨ]
* widzę:
[v i d͡z ɛ]
* rdzą:
[r d͡z ɔ̃]
Occurrences:
6,291
Examples:
* pięć:
[p j ɛ ɲ t͡ɕ]
* chać.:
[x a t͡ɕ]
* ująć:
[u j ɔ ɲ t͡ɕ]
* leci:
[l ɛ t͡ɕ i]
Occurrences:
2,232
Examples:
* dziś.:
[d͡ʑ i ɕ]
* bądź,:
[b ɔ ɲ d͡ʑ]
* "idź,:
[i d͡ʑ]
* dziad:
[d͡ʑ a t]

Sibilant
Occurrences:
11,898
Examples:
* oraz:
[ɔ r a s]
* kasę:
[k a s ɛ]
* stosu:
[s t ɔ s u]
* hałas:
[x a w a s]
Occurrences:
7,808
Examples:
* bluzę:
[b l u z ɛ]
* zyski:
[z ɨ s kʲ i]
* mózgu:
[m u z ɡ u]
* izby:
[i z b ɨ]
Occurrences:
12,832
Examples:
* muszą:
[m u ʂ ɔ̃]
* mysz,:
[m ɨ ʂ]
* czaił:
[t ʂ a j w]
* koszt:
[k ɔ ʂ t]
Occurrences:
4,477
Examples:
* drzwi:
[d ʐ v i]
* też.:
[t ɛ ʐ]
* dużej:
[d u ʐ ɛ j]
* dużo,:
[d u ʐ ɔ]
Occurrences:
4,914
Examples:
* wsi!:
[f ɕ i]
* myśli:
[m ɨ ɕ l i]
* pomyś:
[p ɔ m ɨ ɕ]
* kimś.:
[kʲ i m ɕ]
Occurrences:
595
Examples:
* źle,:
[ʑ l ɛ]
* józi.:
[j u ʑ i]
* bzik:
[b ʑ i k]
* wziął:
[v ʑ ɔ w]

Fricative
Occurrences:
4,910
Examples:
* wsi!:
[f ɕ i]
* twój!:
[t f u j]
* film:
[f i l m]
* świt:
[ɕ f i t]
Occurrences:
14,809
Examples:
* pewny:
[p ɛ v n ɨ]
* drzwi:
[d ʐ v i]
* ewą:
[ɛ v ɔ̃]
* piwa:
[p i v a]

Approximant
Occurrences:
11,658
Examples:
* brał:
[b r a w]
* nił:
[ɲ i w]
* płód.:
[p w u d]
* hałas:
[x a w a s]
Occurrences:
14,074
Examples:
* "jaką:
[j a k ɔ̃]
* pięć:
[p j ɛ ɲ t͡ɕ]
* twój!:
[t f u j]
* czaił:
[t ʂ a j w]

Trill
Occurrences:
14,846
Examples:
* król:
[k r u l]
* oraz:
[ɔ r a s]
* brał:
[b r a w]
* ryby.:
[r ɨ b ɨ]

Lateral
Occurrences:
8,941
Examples:
* król:
[k r u l]
* luba;:
[l u b a]
* myśli:
[m ɨ ɕ l i]
* bluzę:
[b l u z ɛ]

Vowels#

Vowel symbols to the left of are unrounded and those to the right are rounded.

Oral Vowels#

Front

Near-Front

Central

Near-Back

Back

Close
Occurrences:
14,533
Examples:
* wsi!:
[f ɕ i]
* kim,:
[kʲ i m]
* nił:
[ɲ i w]
* drzwi:
[d ʐ v i]
Occurrences:
17,684
Examples:
* pewny:
[p ɛ v n ɨ]
* ryby.:
[r ɨ b ɨ]
* mysz,:
[m ɨ ʂ]
* myśli:
[m ɨ ɕ l i]
Occurrences:
12,946
Examples:
* król:
[k r u l]
* ud.:
[u d]
* płód.:
[p w u d]
* stosu:
[s t ɔ s u]

Close-Mid

Open-Mid
Occurrences:
33,501
Examples:
* kasę:
[k a s ɛ]
* pewny:
[p ɛ v n ɨ]
* pięć:
[p j ɛ ɲ t͡ɕ]
* też.:
[t ɛ ʐ]
Occurrences:
33,574
Examples:
* oraz:
[ɔ r a s]
* stosu:
[s t ɔ s u]
* pomyś:
[p ɔ m ɨ ɕ]
* molo.:
[m ɔ l ɔ]

Open
Occurrences:
38,405
Examples:
* oraz:
[ɔ r a s]
* brał:
[b r a w]
* kasę:
[k a s ɛ]
* "jaką:
[j a k ɔ̃]

Nasal Vowels#

Front

Near-Front

Central

Near-Back

Back

Close

Close-Mid

Open-Mid
Occurrences:
1,455
Examples:
* się”:
[ɕ ɛ̃]
* rękę,:
[r ɛ ŋ k ɛ̃]
* minę!:
[m i n ɛ̃]
* męża:
[m ɛ̃ ʐ a]
Occurrences:
2,717
Examples:
* "jaką:
[j a k ɔ̃]
* muszą:
[m u ʂ ɔ̃]
* ewą:
[ɛ v ɔ̃]
* tobą.:
[t ɔ b ɔ̃]

Open
On this page
Show Source