Ctrl+K
Ctrl+K

Czech CV dictionary v2.0.0#

  • Maintainer: Vox Communis

  • Language: Czech

  • Dialect: N/A

  • Phone set: XPF

  • Number of words: 35,296

  • Phones: a au auː aː b c d f g i iː j k l m n o ou oː p r r̝ s t ts tʃ u uː v x z ŋ ɛ ɛu ɛuː ɛː ɟ ɡ ɦ ɲ ʃ ʒ

  • 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 czech_cv

Or download from the release page.

Intended use#

This dictionary is intended for forced alignment of Czech transcripts.

This dictionary uses the XPF phone set for Czech, and was used in training the Czech XPF 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

Alveopalatal

Palatal

Velar

Glottal

Nasal
Occurrences:
8,701
Examples:
* dodám:
[d o d aː m]
* mění:
[m j ɛ ɲ iː]
* němec:
[ɲ ɛ m ɛ ts]
* nemá:
[n ɛ m aː]
Occurrences:
12,638
Examples:
* plno:
[p l n o]
* nemá:
[n ɛ m aː]
* roman:
[r o m a n]
* naučí:
[n au tʃ iː]
Occurrences:
5,482
Examples:
* mění:
[m j ɛ ɲ iː]
* němec:
[ɲ ɛ m ɛ ts]
* nikdy:
[ɲ i k d i]
* věžní:
[v j ɛ ʒ ɲ iː]
Occurrences:
439
Examples:
* kings:
[k i ŋ g s]
* banka:
[b a ŋ k a]
* range:
[r a ŋ g ɛ]
* kongo:
[k o ŋ g o]

Stop
Occurrences:
9,010
Examples:
* pásu:
[p aː s u]
* posud:
[p o s u d]
* plno:
[p l n o]
* cup:
[ts u p]
Occurrences:
4,062
Examples:
* beze:
[b ɛ z ɛ]
* banka:
[b a ŋ k a]
* byste:
[b i s t ɛ]
* pohyb:
[p o ɦ i b]
Occurrences:
11,408
Examples:
* lišit:
[l i ʃ i t]
* ostrý:
[o s t r iː]
* útlém:
[uː t l ɛː m]
* držet:
[d r ʒ ɛ t]
Occurrences:
7,124
Examples:
* posud:
[p o s u d]
* dodám:
[d o d aː m]
* sídlo:
[s iː d l o]
* nikdy:
[ɲ i k d i]
Occurrences:
2,505
Examples:
* těžší:
[c ɛ ʒ ʃ iː]
* aktiv:
[a k c i v]
* dutě:
[d u c ɛ]
* tik:
[c i k]
Occurrences:
1,568
Examples:
* ledě:
[l ɛ ɟ ɛ]
* oddíl:
[o d ɟ iː l]
* dis:
[ɟ i s]
* dítě:
[ɟ iː c ɛ]
Occurrences:
10,417
Examples:
* kings:
[k i ŋ g s]
* ukážu:
[u k aː ʒ u]
* akcí:
[a k ts iː]
* nikdy:
[ɲ i k d i]
Occurrences:
1,111
Examples:
* gogh:
[ɡ o ɡ ɦ]
* high:
[ɦ i ɡ ɦ]
* page:
[p a ɡ ɛ]
* voksg:
[v o k s ɡ]

Affricate
Occurrences:
4,297
Examples:
* akcí:
[a k ts iː]
* oceli:
[o ts ɛ l i]
* němec:
[ɲ ɛ m ɛ ts]
* hits:
[ɦ i ts]
Occurrences:
2,617
Examples:
* naučí:
[n au tʃ iː]
* číňan:
[tʃ iː ɲ a n]
* čelo:
[tʃ ɛ l o]
* moč:
[m o tʃ]

Sibilant
Occurrences:
11,300
Examples:
* pásu:
[p aː s u]
* kings:
[k i ŋ g s]
* posud:
[p o s u d]
* sídlo:
[s iː d l o]
Occurrences:
5,731
Examples:
* zruš:
[z r u ʃ]
* beze:
[b ɛ z ɛ]
* zimy:
[z i m i]
* fuzzy:
[f u z z i]
Occurrences:
2,149
Examples:
* zruš:
[z r u ʃ]
* šachu:
[ʃ a x u]
* těžší:
[c ɛ ʒ ʃ iː]
* lišit:
[l i ʃ i t]
Occurrences:
1,805
Examples:
* ukážu:
[u k aː ʒ u]
* těžší:
[c ɛ ʒ ʃ iː]
* věžní:
[v j ɛ ʒ ɲ iː]
* žil:
[ʒ i l]

Fricative
Occurrences:
1,100
Examples:
* fuzzy:
[f u z z i]
* kufr:
[k u f r]
* ašraf:
[a ʃ r a f]
* efekt:
[ɛ f ɛ k t]
Occurrences:
11,617
Examples:
* ivy:
[i v i]
* aktiv:
[a k c i v]
* dvoře:
[d v o r̝ ɛ]
* vař:
[v a r̝]
Occurrences:
3,899
Examples:
* gogh:
[ɡ o ɡ ɦ]
* high:
[ɦ i ɡ ɦ]
* hits:
[ɦ i ts]
* výhod:
[v iː ɦ o d]

Approximant
Occurrences:
5,313
Examples:
* mění:
[m j ɛ ɲ iː]
* jdeme:
[j d ɛ m ɛ]
* měna:
[m j ɛ n a]
* věžní:
[v j ɛ ʒ ɲ iː]

Trill
Occurrences:
11,501
Examples:
* zruš:
[z r u ʃ]
* ruka:
[r u k a]
* roman:
[r o m a n]
* ostrý:
[o s t r iː]

Lateral
Occurrences:
12,807
Examples:
* ledě:
[l ɛ ɟ ɛ]
* lišit:
[l i ʃ i t]
* oceli:
[o ts ɛ l i]
* plno:
[p l n o]

Vowels#

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

Front

Near-Front

Central

Near-Back

Back

Close
Occurrences:
17,350
Examples:
* kings:
[k i ŋ g s]
* lišit:
[l i ʃ i t]
* oceli:
[o ts ɛ l i]
* ivy:
[i v i]
Occurrences:
11,085
Examples:
* těžší:
[c ɛ ʒ ʃ iː]
* akcí:
[a k ts iː]
* mění:
[m j ɛ ɲ iː]
* sídlo:
[s iː d l o]
Occurrences:
7,041
Examples:
* zruš:
[z r u ʃ]
* šachu:
[ʃ a x u]
* pásu:
[p aː s u]
* ukážu:
[u k aː ʒ u]
Occurrences:
1,727
Examples:
* útlém:
[uː t l ɛː m]
* pánům:
[p aː n uː m]
* týdnů:
[t iː d n uː]
* členů:
[tʃ l ɛ n uː]

Close-Mid
Occurrences:
19,779
Examples:
* gogh:
[ɡ o ɡ ɦ]
* posud:
[p o s u d]
* dodám:
[d o d aː m]
* oceli:
[o ts ɛ l i]
Occurrences:
108
Examples:
* tóra:
[t oː r a]
* miró:
[m i r oː]
* león:
[l ɛ oː n]
* fóra:
[f oː r a]

Open-Mid
Occurrences:
22,011
Examples:
* ledě:
[l ɛ ɟ ɛ]
* těžší:
[c ɛ ʒ ʃ iː]
* oceli:
[o ts ɛ l i]
* mění:
[m j ɛ ɲ iː]
Occurrences:
3,298
Examples:
* útlém:
[uː t l ɛː m]
* švédy:
[ʃ v ɛː d i]
* slané:
[s l a n ɛː]
* levé:
[l ɛ v ɛː]

Open
Occurrences:
17,365
Examples:
* šachu:
[ʃ a x u]
* akcí:
[a k ts iː]
* aktiv:
[a k c i v]
* ruka:
[r u k a]
Occurrences:
6,747
Examples:
* pásu:
[p aː s u]
* ukážu:
[u k aː ʒ u]
* dodám:
[d o d aː m]
* nemá:
[n ɛ m aː]

Diphthongs#

  • au

  • auː

  • ou

  • ɛu

  • ɛuː

On this page
Show Source