Ctrl+K
Ctrl+K

Tatar CV dictionary v2.0.0#

  • Maintainer: Vox Communis

  • Language: Tatar

  • Dialect: N/A

  • Phone set: Epitran

  • Number of words: 22,664

  • Phones: a b d e f h i j k l m n o p r s t u w x y z ø ŋ ɑ ɕ ɡ ɤ ɯ ʃ ʑ ʒ ʔ

  • 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 tatar_cv

Or download from the release page.

Intended use#

This dictionary is intended for forced alignment of Tatar transcripts.

This dictionary uses the Epitran phone set for Tatar, and was used in training the Tatar 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

Alveopalatal

Palatal

Velar

Glottal

Nasal
Occurrences:
5,846
Examples:
* сумга:
[s u m ɡ ɑ]
* кидем:
[k i d e m]
* форум:
[f o r u m]
* миләш:
[m i l a ʃ]
Occurrences:
12,539
Examples:
* җыен:
[ʑ ɤ e n]
* кенә:
[k e n a]
* туена:
[t u e n ɑ]
* иркен:
[i r k e n]
Occurrences:
2,356
Examples:
* ирең:
[i r e ŋ]
* туң:
[t u ŋ]
* сукаң:
[s u k ɑ ŋ]
* аеның:
[ɑ e n ɤ ŋ]

Stop
Occurrences:
2,019
Examples:
* торып:
[t o r ɤ p]
* кисеп:
[k i s e p]
* биләп:
[b i l a p]
* апам:
[ɑ p ɑ m]
Occurrences:
3,521
Examples:
* буаз:
[b u ɑ z]
* биләп:
[b i l a p]
* бите:
[b i t e]
* бәһа:
[b a h ɑ]
Occurrences:
8,955
Examples:
* уятты:
[u j ɑ t t ɤ]
* туена:
[t u e n ɑ]
* туй:
[t u j]
* торып:
[t o r ɤ p]
Occurrences:
4,150
Examples:
* радар:
[r ɑ d ɑ r]
* дияр:
[d i j ɑ r]
* кидем:
[k i d e m]
* чыдар:
[ɕ ɤ d ɑ r]
Occurrences:
9,337
Examples:
* ачлык:
[ɑ ɕ l ɤ k]
* кенә:
[k e n a]
* торак:
[t o r ɑ k]
* кидем:
[k i d e m]
Occurrences:
5,630
Examples:
* сумга:
[s u m ɡ ɑ]
* азагы:
[ɑ z ɑ ɡ ɤ]
* ачуга:
[ɑ ɕ u ɡ ɑ]
* ягъни:
[j ɑ ɡ ʔ n i]
Occurrences:
577
Examples:
* ягъни:
[j ɑ ɡ ʔ n i]
* июль:
[i ɯ l ʔ]
* тверь:
[t w e r ʔ]
* яшьне:
[j ɑ ʃ ʔ n e]

Sibilant
Occurrences:
5,631
Examples:
* сумга:
[s u m ɡ ɑ]
* кисеп:
[k i s e p]
* саумы:
[s ɑ u m ɤ]
* аскын:
[ɑ s k ɤ n]
Occurrences:
2,840
Examples:
* буаз:
[b u ɑ z]
* азагы:
[ɑ z ɑ ɡ ɤ]
* йөзә:
[j ø z a]
* зират:
[z i r ɑ t]
Occurrences:
3,051
Examples:
* эшләү:
[e ʃ l a y]
* миләш:
[m i l a ʃ]
* кашык:
[k ɑ ʃ ɤ k]
* кыш:
[k ɤ ʃ]
Occurrences:
112
Examples:
* пляж:
[p l j ɑ ʒ]
* жгут:
[ʒ ɡ u t]
* жуаны:
[ʒ u ɑ n ɤ]
* стаж:
[s t ɑ ʒ]
Occurrences:
2,823
Examples:
* ачлык:
[ɑ ɕ l ɤ k]
* ачуга:
[ɑ ɕ u ɡ ɑ]
* чыдар:
[ɕ ɤ d ɑ r]
* мичкә:
[m i ɕ k a]
Occurrences:
863
Examples:
* җыен:
[ʑ ɤ e n]
* әҗере:
[a ʑ e r e]
* җаным:
[ʑ ɑ n ɤ m]
* җигеп:
[ʑ i ɡ e p]

Fricative
Occurrences:
598
Examples:
* форум:
[f o r u m]
* фани:
[f ɑ n i]
* туфан:
[t u f ɑ n]
* җәфа:
[ʑ a f ɑ]
Occurrences:
291
Examples:
* бәһа:
[b a h ɑ]
* шәһри:
[ʃ a h r i]
* мөһим:
[m ø h i m]
* һөнәр:
[h ø n a r]

Approximant
Occurrences:
937
Examples:
* вобла:
[w o b l ɑ]
* ватып:
[w ɑ t ɤ p]
* тавис:
[t ɑ w i s]
* явым:
[j ɑ w ɤ m]
Occurrences:
4,552
Examples:
* уятты:
[u j ɑ t t ɤ]
* туй:
[t u j]
* дияр:
[d i j ɑ r]
* укый:
[u k ɤ j]

Trill
Occurrences:
12,208
Examples:
* торып:
[t o r ɤ p]
* торак:
[t o r ɑ k]
* радар:
[r ɑ d ɑ r]
* әҗере:
[a ʑ e r e]

Lateral
Occurrences:
10,913
Examples:
* эшләү:
[e ʃ l a y]
* ачлык:
[ɑ ɕ l ɤ k]
* килен:
[k i l e n]
* миләш:
[m i l a ʃ]

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:
5,455
Examples:
* дияр:
[d i j ɑ r]
* кидем:
[k i d e m]
* кисеп:
[k i s e p]
* иркен:
[i r k e n]
Occurrences:
1,884
Examples:
* эшләү:
[e ʃ l a y]
* үлем:
[y l e m]
* сөрү:
[s ø r y]
* күн:
[k y n]
Occurrences:
311
Examples:
* июль:
[i ɯ l ʔ]
* бюст:
[b ɯ s t]
* юка:
[ɯ k ɑ]
* ешаю:
[j e ʃ ɑ ɯ]
Occurrences:
3,554
Examples:
* буаз:
[b u ɑ z]
* уятты:
[u j ɑ t t ɤ]
* туена:
[t u e n ɑ]
* туй:
[t u j]

Close-Mid
Occurrences:
12,038
Examples:
* эшләү:
[e ʃ l a y]
* җыен:
[ʑ ɤ e n]
* кенә:
[k e n a]
* туена:
[t u e n ɑ]
Occurrences:
1,664
Examples:
* йөзә:
[j ø z a]
* йөзе:
[j ø z e]
* өчен:
[ø ɕ e n]
* сөрү:
[s ø r y]
Occurrences:
12,544
Examples:
* җыен:
[ʑ ɤ e n]
* ачлык:
[ɑ ɕ l ɤ k]
* уятты:
[u j ɑ t t ɤ]
* торып:
[t o r ɤ p]
Occurrences:
2,507
Examples:
* торып:
[t o r ɤ p]
* торак:
[t o r ɑ k]
* форум:
[f o r u m]
* йокым:
[j o k ɤ m]

Open-Mid

Open
Occurrences:
11,405
Examples:
* эшләү:
[e ʃ l a y]
* кенә:
[k e n a]
* әҗере:
[a ʑ e r e]
* йөзә:
[j ø z a]
Occurrences:
21,754
Examples:
* ачлык:
[ɑ ɕ l ɤ k]
* буаз:
[b u ɑ z]
* уятты:
[u j ɑ t t ɤ]
* туена:
[t u e n ɑ]
On this page
Show Source