Ctrl+K
Ctrl+K

Uyghur CV dictionary v2.0.0#

  • Maintainer: Vox Communis

  • Language: Uyghur

  • Dialect: N/A

  • Phone set: Epitran

  • Number of words: 23,899

  • Phones: a b d d͡ʒ e f h i j k l m n o p q r s t t͡ʃ u w y z ø ŋ ɛ ɡ ʁ ʃ ʒ χ

  • 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 uyghur_cv

Or download from the release page.

Intended use#

This dictionary is intended for forced alignment of Uyghur transcripts.

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

Uvular

Glottal

Nasal
Occurrences:
7,434
Examples:
* مانجۇ:
[m a n d͡ʒ u]
* مۇنۇ:
[m u n u]
* ئاتام:
[a t a m]
* مېتال:
[m e t a l]
Occurrences:
11,429
Examples:
* تىكەن:
[t i k ɛ n]
* تەنگە:
[t ɛ n ɡ ɛ]
* ئورۇن:
[o r u n]
* نەپەر:
[n ɛ p ɛ r]
Occurrences:
3,064
Examples:
* ئاڭغا:
[a ŋ ʁ a]
* كالاڭ:
[k a l a ŋ]
* جەڭدە:
[d͡ʒ ɛ ŋ d ɛ]
* ئوڭ:
[o ŋ]

Stop
Occurrences:
3,575
Examples:
* شەپەق:
[ʃ ɛ p ɛ q]
* نەپەر:
[n ɛ p ɛ r]
* ئۇلاپ:
[u l a p]
* پولات:
[p o l a t]
Occurrences:
2,680
Examples:
* بۈيۈك:
[b y j y k]
* بىخۇد:
[b i χ u d]
* كەبى:
[k ɛ b i]
* غايىب:
[ʁ a j i b]
Occurrences:
9,288
Examples:
* تىكەن:
[t i k ɛ n]
* تەخسە:
[t ɛ χ s ɛ]
* تەنگە:
[t ɛ n ɡ ɛ]
* تۇتاش:
[t u t a ʃ]
Occurrences:
6,616
Examples:
* ھىدى،:
[h i d i]
* بىخۇد:
[b i χ u d]
* جايدا:
[d͡ʒ a j d a]
* دىن:
[d i n]
Occurrences:
5,754
Examples:
* تىكەن:
[t i k ɛ n]
* بۈيۈك:
[b y j y k]
* كەلسە:
[k ɛ l s ɛ]
* كوزا:
[k o z a]
Occurrences:
1,998
Examples:
* تەنگە:
[t ɛ n ɡ ɛ]
* سۆگەت:
[s ø ɡ ɛ t]
* ئىگە:
[i ɡ ɛ]
* گۆھەر:
[ɡ ø h ɛ r]
Occurrences:
7,051
Examples:
* شەپەق:
[ʃ ɛ p ɛ q]
* قۇر:
[q u r]
* يىراق:
[j i r a q]
* توقۇش:
[t o q u ʃ]

Affricate
Occurrences:
2,769
Examples:
* چىش:
[t͡ʃ i ʃ]
* چېچىپ:
[t͡ʃ e t͡ʃ i p]
* چىراغ:
[t͡ʃ i r a ʁ]
* چۆمۈچ:
[t͡ʃ ø m y t͡ʃ]
Occurrences:
675
Examples:
* مانجۇ:
[m a n d͡ʒ u]
* جايدا:
[d͡ʒ a j d a]
* جىم:
[d͡ʒ i m]
* جەڭدە:
[d͡ʒ ɛ ŋ d ɛ]

Sibilant
Occurrences:
5,063
Examples:
* تەخسە:
[t ɛ χ s ɛ]
* كەلسە:
[k ɛ l s ɛ]
* سىر:
[s i r]
* سۆگەت:
[s ø ɡ ɛ t]
Occurrences:
2,967
Examples:
* ئەزا:
[ɛ z a]
* غەرەز:
[ʁ ɛ r ɛ z]
* كوزا:
[k o z a]
* ۋەزىر:
[w ɛ z i r]
Occurrences:
3,861
Examples:
* ئىشلە:
[i ʃ l ɛ]
* شەپەق:
[ʃ ɛ p ɛ q]
* تۇتاش:
[t u t a ʃ]
* چىش:
[t͡ʃ i ʃ]
Occurrences:
23
Examples:
* پارىژ:
[p a r i ʒ]

Fricative
Occurrences:
78
Examples:
* نېفىت:
[n e f i t]
* فىلىم:
[f i l i m]
* فوتو:
[f o t o]
* لىفت:
[l i f t]
Occurrences:
1,024
Examples:
* تەخسە:
[t ɛ χ s ɛ]
* تېخى.:
[t e χ i]
* خەش:
[χ ɛ ʃ]
* خىيال:
[χ i j a l]
Occurrences:
3,254
Examples:
* ئاڭغا:
[a ŋ ʁ a]
* غەرەز:
[ʁ ɛ r ɛ z]
* نانغا:
[n a n ʁ a]
* غايىب:
[ʁ a j i b]
Occurrences:
1,293
Examples:
* ھۆر:
[h ø r]
* ھىدى،:
[h i d i]
* ھۆكۈم:
[h ø k y m]
* ھەددى:
[h ɛ d d i]

Approximant
Occurrences:
1,730
Examples:
* تەۋە:
[t ɛ w ɛ]
* ۋاي:
[w a j]
* تۆۋە:
[t ø w ɛ]
* ۋەزىر:
[w ɛ z i r]
Occurrences:
5,722
Examples:
* ۋاي:
[w a j]
* بۈيۈك:
[b y j y k]
* خىيال:
[χ i j a l]
* ياسا.:
[j a s a]

Trill
Occurrences:
9,266
Examples:
* ئورۇن:
[o r u n]
* نەپەر:
[n ɛ p ɛ r]
* سىر:
[s i r]
* ھۆر:
[h ø r]

Lateral
Occurrences:
11,486
Examples:
* ئىشلە:
[i ʃ l ɛ]
* ئەل:
[ɛ l]
* كەلسە:
[k ɛ l s ɛ]
* نۆل:
[n ø l]

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:
29,982
Examples:
* ئىشلە:
[i ʃ l ɛ]
* تىكەن:
[t i k ɛ n]
* چىش:
[t͡ʃ i ʃ]
* تېخى.:
[t e χ i]
Occurrences:
2,821
Examples:
* بۈيۈك:
[b y j y k]
* ئۆلۈش:
[ø l y ʃ]
* ھۆكۈم:
[h ø k y m]
* چۆمۈچ:
[t͡ʃ ø m y t͡ʃ]
Occurrences:
7,311
Examples:
* تۇتاش:
[t u t a ʃ]
* ئورۇن:
[o r u n]
* مانجۇ:
[m a n d͡ʒ u]
* مۇنۇ:
[m u n u]

Close-Mid
Occurrences:
3,060
Examples:
* تېخى.:
[t e χ i]
* مېتال:
[m e t a l]
* ئېتىش:
[e t i ʃ]
* چېچىپ:
[t͡ʃ e t͡ʃ i p]
Occurrences:
1,747
Examples:
* ھۆر:
[h ø r]
* نۆل:
[n ø l]
* ئۆلۈش:
[ø l y ʃ]
* سۆگەت:
[s ø ɡ ɛ t]
Occurrences:
3,277
Examples:
* ئورۇن:
[o r u n]
* كوزا:
[k o z a]
* پولات:
[p o l a t]
* توقۇش:
[t o q u ʃ]

Open-Mid
Occurrences:
11,320
Examples:
* ئىشلە:
[i ʃ l ɛ]
* تىكەن:
[t i k ɛ n]
* تەخسە:
[t ɛ χ s ɛ]
* شەپەق:
[ʃ ɛ p ɛ q]

Open
Occurrences:
17,947
Examples:
* تۇتاش:
[t u t a ʃ]
* ۋاي:
[w a j]
* ئەزا:
[ɛ z a]
* ئاڭغا:
[a ŋ ʁ a]
On this page
Show Source