Ctrl+K
Ctrl+K

Armenian CV dictionary v2.0.0#

  • Maintainer: Vox Communis

  • Language: Armenian

  • Dialect: N/A

  • Phone set: XPF

  • Number of words: 3,282

  • Phones: b d f h i j k kʰ l m n p pʰ r s sʰ t tʰ u v z ɑ ɔ ə ɛ ɡ ɾ ʁ ʃ ʃʰ 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 armenian_cv

Or download from the release page.

Intended use#

This dictionary is intended for forced alignment of Armenian transcripts.

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

Glottal

Nasal
Occurrences:
1,108
Examples:
* մեջ:
[m ɛ d ʒ]
* մայր:
[m ɑ j ɾ]
* դեմ:
[d ɛ m]
* սեմը:
[s ɛ m ə]
Occurrences:
2,732
Examples:
* ինը:
[i n ə]
* առանց:
[ɑ r ɑ n t sʰ]
* գոլին:
[ɡ ɔ l i n]
* քունի:
[kʰ u n i]

Stop
Occurrences:
321
Examples:
* պէտք:
[p ɛ t kʰ]
* կապել:
[k ɑ p ɛ l]
* որպես:
[ɔ ɾ p ɛ s]
* կապով:
[k ɑ p ɔ v]
Occurrences:
345
Examples:
* բրնձի:
[b ɾ n d z i]
* ժաբոն:
[ʒ ɑ b ɔ n]
* բծեր:
[b t s ɛ ɾ]
* բութ:
[b u tʰ]
Occurrences:
1,878
Examples:
* կից:
[k i t sʰ]
* ցոփ:
[t sʰ ɔ pʰ]
* առանց:
[ɑ r ɑ n t sʰ]
* պէտք:
[p ɛ t kʰ]
Occurrences:
641
Examples:
* մեջ:
[m ɛ d ʒ]
* ձախ:
[d z ɑ χ]
* խնդրի:
[χ n d ɾ i]
* դեմ:
[d ɛ m]
Occurrences:
1,026
Examples:
* կից:
[k i t sʰ]
* կարեց:
[k ɑ ɾ ɛ t sʰ]
* մաշկ:
[m ɑ ʃ k]
* տակ՝:
[t ɑ k]
Occurrences:
415
Examples:
* գոլին:
[ɡ ɔ l i n]
* գծեր:
[ɡ t s ɛ ɾ]
* գլուխ:
[ɡ l u χ]
* գնման:
[ɡ n m ɑ n]

Sibilant Plain
Occurrences:
924
Examples:
* գծեր:
[ɡ t s ɛ ɾ]
* օրս:
[ɔ ɾ s]
* սեմը:
[s ɛ m ə]
* տուած:
[t u ɑ t s]
Occurrences:
345
Examples:
* ձախ:
[d z ɑ χ]
* բրնձի:
[b ɾ n d z i]
* զգալի:
[z ɡ ɑ l i]
* զենքը:
[z ɛ n kʰ ə]
Occurrences:
324
Examples:
* շան:
[ʃ ɑ n]
* մաշկ:
[m ɑ ʃ k]
* շվայտ:
[ʃ v ɑ j t]
* ուշ:
[u ʃ]
Occurrences:
226
Examples:
* մեջ:
[m ɛ d ʒ]
* առջևի:
[ɑ r d ʒ ɛ v i]
* ժաբոն:
[ʒ ɑ b ɔ n]
* ողջ:
[ɔ ʁ d ʒ]

Aspirated
Occurrences:
478
Examples:
* կից:
[k i t sʰ]
* ցոփ:
[t sʰ ɔ pʰ]
* առանց:
[ɑ r ɑ n t sʰ]
* կարեց:
[k ɑ ɾ ɛ t sʰ]
Occurrences:
121
Examples:
* մինչև:
[m i n t ʃʰ ɛ v]
* աչքեր:
[ɑ t ʃʰ kʰ ɛ ɾ]
* չափից:
[t ʃʰ ɑ pʰ i t sʰ]
* գրչին:
[ɡ ɾ t ʃʰ i n]

Fricative
Occurrences:
46
Examples:
* ֆոլքի:
[f ɔ l kʰ i]
* ֆոնդը:
[f ɔ n d ə]
* ֆեռալ:
[f ɛ r ɑ l]
Occurrences:
773
Examples:
* մինչև:
[m i n t ʃʰ ɛ v]
* հովիտ:
[h ɔ v i t]
* տվել:
[t v ɛ l]
* անվան:
[ɑ n v ɑ n]
Occurrences:
195
Examples:
* ձախ:
[d z ɑ χ]
* խնդրի:
[χ n d ɾ i]
* գլուխ:
[ɡ l u χ]
* թխել:
[tʰ χ ɛ l]
Occurrences:
329
Examples:
* ցեղը:
[t sʰ ɛ ʁ ə]
* ուղիղ:
[u ʁ i ʁ]
* ողջ:
[ɔ ʁ d ʒ]
* դեղին:
[d ɛ ʁ i n]
Occurrences:
476
Examples:
* հիմար:
[h i m ɑ ɾ]
* հովիտ:
[h ɔ v i t]
* հնարք:
[h n ɑ ɾ kʰ]
* հարկ:
[h ɑ ɾ k]

Approximant
Occurrences:
714
Examples:
* մայր:
[m ɑ j ɾ]
* այն:
[ɑ j n]
* այսօր:
[ɑ j s ɔ ɾ]
* լայն:
[l ɑ j n]

Tap
Occurrences:
2,246
Examples:
* արքա:
[ɑ ɾ kʰ ɑ]
* գծեր:
[ɡ t s ɛ ɾ]
* արուն:
[ɑ ɾ u n]
* որն:
[ɔ ɾ n]

Trill
Occurrences:
193
Examples:
* առանց:
[ɑ r ɑ n t sʰ]
* անսեռ:
[ɑ n s ɛ r]
* ռուս:
[r u s]
* առջևի:
[ɑ r d ʒ ɛ v i]

Lateral
Occurrences:
523
Examples:
* գոլին:
[ɡ ɔ l i n]
* գլուխ:
[ɡ l u χ]
* թխել:
[tʰ χ ɛ l]
* տվել:
[t v ɛ 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:
1,456
Examples:
* կից:
[k i t sʰ]
* ինը:
[i n ə]
* գոլին:
[ɡ ɔ l i n]
* քունի:
[kʰ u n i]
Occurrences:
1,310
Examples:
* քունի:
[kʰ u n i]
* արուն:
[ɑ ɾ u n]
* թուրք:
[tʰ u ɾ kʰ]
* տուած:
[t u ɑ t s]

Close-Mid
Occurrences:
478
Examples:
* ինը:
[i n ə]
* սեմը:
[s ɛ m ə]
* ծծանը:
[t s t s ɑ n ə]
* ցեղը:
[t sʰ ɛ ʁ ə]

Open-Mid
Occurrences:
1,922
Examples:
* մեջ:
[m ɛ d ʒ]
* պէտք:
[p ɛ t kʰ]
* գծեր:
[ɡ t s ɛ ɾ]
* դեմ:
[d ɛ m]
Occurrences:
909
Examples:
* ցոփ:
[t sʰ ɔ pʰ]
* գոլին:
[ɡ ɔ l i n]
* որն:
[ɔ ɾ n]
* անոր:
[ɑ n ɔ ɾ]

Open
Occurrences:
4,463
Examples:
* առանց:
[ɑ r ɑ n t sʰ]
* ձախ:
[d z ɑ χ]
* արքա:
[ɑ ɾ kʰ ɑ]
* արուն:
[ɑ ɾ u n]
On this page
Show Source