Ctrl+K
Ctrl+K

Abkhaz CV dictionary v2.0.0#

  • Maintainer: Vox Communis

  • Language: Abkhaz

  • Dialect: N/A

  • Phone set: XPF

  • Number of words: 3,671

  • Phones: a b d dz dʒʲ dʷ f j k kʲ kʲʼ kʷ kʷʼ kʼ l m n p pʼ qʲʼ qʷʼ qʼ r s t ts tsʼ tʃ tʃʲ tʃʲʼ tʃʷ tʃʼ tʷ tʷʼ tʼ v w z ħ ħʷ ə ɡ ɡʲ ɡʷ ʁ ʁʲ ʁʷ ʃ ʃʲ ʃʷ ʒ ʒʲ ʒʷ ʡ χ χʲ χʷ

  • 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 abkhaz_cv

Or download from the release page.

Intended use#

This dictionary is intended for forced alignment of Abkhaz transcripts.

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

Uvular

Pharyngeal

Epiglottal

Nasal
Occurrences:
639
Examples:
* амши:
[a m ʃ j]
* адам:
[a d a m]
* сумуд:
[s w m w d]
* аимак:
[a j m a kʼ]
Occurrences:
850
Examples:
* нцәак:
[n ts w a kʼ]
* рыцны:
[r ə ts n ə]
* ааины:
[a a j n ə]
* хынтә:
[χ ə n tʷʼ]

Stop
Occurrences:
405
Examples:
* алԥха:
[a l p χ a]
* аԥхыӡ:
[a p χ ə dz]
* уԥсны:
[w p s n ə]
* аҳаԥы:
[a ħ a p ə]
Occurrences:
329
Examples:
* убас:
[w b a s]
* иаби:
[j a b j]
* дбаны:
[d b a n ə]
* абза:
[a b z a]
Occurrences:
473
Examples:
* рызҭо:
[r ə z t a]
* ухаҭа:
[w χ a t a]
* еиҭах:
[a j t a χ]
* рышәҭ:
[r ə ʃʷ t]
Occurrences:
432
Examples:
* адам:
[a d a m]
* дауҭ:
[d a w t]
* сумуд:
[s w m w d]
* дҵас:
[d tsʼ a s]
Occurrences:
3
Examples:
* ақды:
[a k d ə]
Occurrences:
265
Examples:
* угыл:
[w ɡ ə l]
* грак:
[ɡ r a kʼ]
* ижәга:
[j ʒʷ ɡ a]
* иааг:
[j a a ɡ]
Occurrences:
228
Examples:
* ҩызас:
[ʡ ə z a s]
* ҩызак:
[ʡ ə z a kʼ]
* ҩба:
[ʡ b a]
* ихшыҩ:
[j χ ʃ ə ʡ]

Affricate
Occurrences:
410
Examples:
* нцәак:
[n ts w a kʼ]
* иацы:
[j a ts ə]
* рыцны:
[r ə ts n ə]
* ацәа:
[a ts w a]
Occurrences:
285
Examples:
* аӡә:
[a dz w]
* аԥхыӡ:
[a p χ ə dz]
* ауӡом:
[a w dz a m]
* зынӡа:
[z ə n dz a]
Occurrences:
62
Examples:
* амахҽ:
[a m a χ tʃ]
* аҽны:
[a tʃ n ə]
* даҽа:
[d a tʃ a]
* иҽхәо:
[j tʃ χʷ a]

Sibilant
Occurrences:
373
Examples:
* убас:
[w b a s]
* ҩызас:
[ʡ ə z a s]
* сахь:
[s a χʲ]
* ҭысуа:
[t ə s w a]
Occurrences:
1,016
Examples:
* рызҭо:
[r ə z t a]
* хазҵо:
[χ a z tsʼ a]
* ҩызас:
[ʡ ə z a s]
* рзгьы:
[r z ɡʲ ə]
Occurrences:
384
Examples:
* иаша:
[j a ʃ a]
* амши:
[a m ʃ j]
* мшуп:
[m ʃ w pʼ]
* ахши:
[a χ ʃ j]
Occurrences:
168
Examples:
* шәшьы:
[ʃʷ ʃʲ ə]
* шьахә:
[ʃʲ a χʷ]
* сашьа:
[s a ʃʲ a]
* ишьит:
[j ʃʲ j tʼ]
Occurrences:
537
Examples:
* шәшьы:
[ʃʷ ʃʲ ə]
* рышәҭ:
[r ə ʃʷ t]
* шәгәы:
[ʃʷ ɡʷ ə]
* шәарҭ:
[ʃʷ a r t]
Occurrences:
7
Examples:
* абжа:
[a b ʒ a]
Occurrences:
64
Examples:
* бжьык:
[b ʒʲ ə kʼ]
* абжьы:
[a b ʒʲ ə]
Occurrences:
108
Examples:
* ражәа:
[r a ʒʷ a]
* иажәа:
[j a ʒʷ a]
* уажәа:
[w a ʒʷ a]
* зажәа:
[z a ʒʷ a]

Fricative
Occurrences:
17
Examples:
* ишәфа:
[j ʃʷ f a]
* иусуф:
[j w s w f]
* фба:
[f b a]
Occurrences:
8
Examples:
Occurrences:
607
Examples:
* алԥха:
[a l p χ a]
* хазҵо:
[χ a z tsʼ a]
* ухаҭа:
[w χ a t a]
* еиҭах:
[a j t a χ]
Occurrences:
207
Examples:
* сахь:
[s a χʲ]
* ахь:
[a χʲ]
* ихьӡ:
[j χʲ dz]
* уахь:
[w a χʲ]
Occurrences:
146
Examples:
* шьахә:
[ʃʲ a χʷ]
* арахә:
[a r a χʷ]
* сыхә:
[s ə χʷ]
* хәҭа:
[χʷ t a]
Occurrences:
36
Examples:
* ргәаӷ:
[r ɡʷ a ʁ]
* аӷари:
[a ʁ a r j]
* аӷра:
[a ʁ r a]
* ҳаӷоу:
[ħ a ʁ a w]
Occurrences:
22
Examples:
Occurrences:
20
Examples:
Occurrences:
392
Examples:
* ҳауит:
[ħ a w j tʼ]
* раҳра:
[r a ħ r a]
* иҳаҭ:
[j ħ a t]
* аллаҳ:
[a l l a ħ]
Occurrences:
180
Examples:
* ҳәа-:
[ħʷ a]
* аҳәаа:
[a ħʷ a a]
* аҳәса:
[a ħʷ s a]
* ҳәаа:
[ħʷ a a]

Approximant
Occurrences:
1,174
Examples:
* убас:
[w b a s]
* ухаҭа:
[w χ a t a]
* ауп:
[a w pʼ]
* аӡә:
[a dz w]
Occurrences:
2,388
Examples:
* иаша:
[j a ʃ a]
* амши:
[a m ʃ j]
* еиҭах:
[a j t a χ]
* иаби:
[j a b j]

Trill
Occurrences:
1,749
Examples:
* рызҭо:
[r ə z t a]
* рышәҭ:
[r ə ʃʷ t]
* рзгьы:
[r z ɡʲ ə]
* ргәаӷ:
[r ɡʷ a ʁ]

Lateral
Occurrences:
584
Examples:
* алԥха:
[a l p χ a]
* ахала:
[a χ a l a]
* ила:
[j l a]
* лхы:
[l χ ə]

Vowels#

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

Front

Near-Front

Central

Near-Back

Back

Close

Close-Mid
Occurrences:
2,392
Examples:
* рызҭо:
[r ə z t a]
* шәшьы:
[ʃʷ ʃʲ ə]
* ҩызас:
[ʡ ə z a s]
* рышәҭ:
[r ə ʃʷ t]

Open-Mid

Open
Occurrences:
7,387
Examples:
* рызҭо:
[r ə z t a]
* убас:
[w b a s]
* алԥха:
[a l p χ a]
* иаша:
[j a ʃ a]
On this page
Show Source