Georgian CV dictionary v2.0.0#
@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 georgian_cv
Or download from the release page.
Intended use#
This dictionary is intended for forced alignment of Georgian transcripts.
This dictionary uses the XPF phone set for Georgian, and was used in training the Georgian 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 |
Velar |
Uvular |
Glottal |
---|---|---|---|---|---|---|---|
Nasal |
Occurrences: 4,791 Examples: * ჯამში: [dʒ ɑ m ʃ i] * ძმათა: [dz m ɑ tʰ ɑ] * ემილი: [ɛ m i l i] * ემი: [ɛ m i] |
Occurrences: 4,905 Examples: * ნახონ: [n ɑ x ɔ n] * უნდა: [u n d ɑ] * პრინც: [pʼ ɾ i n ts] * პენი: [pʼ ɛ n i] |
|||||
Stop |
Occurrences: 4,568 Examples: * ღიობი: [ɣ i ɔ b i] * ბზარი: [b z ɑ ɾ i] * ბევრს: [b ɛ v ɾ s] * უბნის: [u b n i s] |
Occurrences: 3,647 Examples: * დადეს: [d ɑ d ɛ s] * უნდა: [u n d ɑ] * დუკი: [d u kʼ i] * უდევს: [u d ɛ v s] |
Occurrences: 2,265 Examples: * გელა: [ɡ ɛ l ɑ] * როგორ: [ɾ ɔ ɡ ɔ ɾ] * მიგელ: [m i ɡ ɛ l] * ედგა: [ɛ d ɡ ɑ] |
||||
Affricate |
Occurrences: 1,167 Examples: * ფიცის: [pʰ i ts i s] * პრინც: [pʼ ɾ i n ts] * როცა: [ɾ ɔ ts ɑ] * არც: [ɑ ɾ ts] Occurrences: 380 Examples: * ძმათა: [dz m ɑ tʰ ɑ] * ძმა: [dz m ɑ] * სიძის: [s i dz i s] * ძირას: [dz i ɾ ɑ s] |
Occurrences: 421 Examples: * კიჩეს: [kʼ i tʃ ɛ s] * დარჩა: [d ɑ ɾ tʃ ɑ] * ჩადის: [tʃ ɑ d i s] * ელჩი: [ɛ l tʃ i] Occurrences: 265 Examples: * ჯამში: [dʒ ɑ m ʃ i] * ჯარის: [dʒ ɑ ɾ i s] * ჯონმა: [dʒ ɔ n m ɑ] * მსაჯი: [m s ɑ dʒ i] |
|||||
Sibilant |
Occurrences: 7,500 Examples: * დადეს: [d ɑ d ɛ s] * ფიცის: [pʰ i ts i s] * ჯარის: [dʒ ɑ ɾ i s] * ფლის: [pʰ l i s] Occurrences: 1,067 Examples: * ზოლს: [z ɔ l s] * ვაზი: [v ɑ z i] * ბზარი: [b z ɑ ɾ i] * ამაზე: [ɑ m ɑ z ɛ] |
Occurrences: 1,769 Examples: * ჯამში: [dʒ ɑ m ʃ i] * პუშტუ: [pʼ u ʃ tʼ u] * ხელში: [x ɛ l ʃ i] * მასში: [m ɑ s ʃ i] Occurrences: 96 Examples: * ანჟი: [ɑ n ʒ i] * ჟანა: [ʒ ɑ n ɑ] * ჟან: [ʒ ɑ n] * ბირჟა: [b i ɾ ʒ ɑ] |
|||||
Fricative |
Occurrences: 3,222 Examples: * იავის: [i ɑ v i s] * ვაზი: [v ɑ z i] * ბევრს: [b ɛ v ɾ s] * უდევს: [u d ɛ v s] |
Occurrences: 179 Examples: * ჰყავთ: [h qʼ ɑ v tʰ] * დაჰყო: [d ɑ h qʼ ɔ] * ჰილს: [h i l s] * ჰავა: [h ɑ v ɑ] |
|||||
Tap |
Occurrences: 6,377 Examples: * ჯარის: [dʒ ɑ ɾ i s] * ერთი: [ɛ ɾ tʰ i] * თორი: [tʰ ɔ ɾ i] * პრინც: [pʼ ɾ i n ts] |
||||||
Lateral |
Occurrences: 5,049 Examples: * ფლის: [pʰ l i s] * ზოლს: [z ɔ l s] * გელა: [ɡ ɛ l ɑ] * ემილი: [ɛ m i l i] |
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: 13,394 Examples: * თაღია: [tʰ ɑ ɣ i ɑ] * ჯამში: [dʒ ɑ m ʃ i] * ფიცის: [pʰ i ts i s] * ჯარის: [dʒ ɑ ɾ i s] |
Occurrences: 3,340 Examples: * პუშტუ: [pʼ u ʃ tʼ u] * უნდა: [u n d ɑ] * დუკი: [d u kʼ i] * ყურის: [qʼ u ɾ i s] |
|||
Close-Mid |
|||||
Open-Mid |
Occurrences: 9,885 Examples: * დადეს: [d ɑ d ɛ s] * ერთი: [ɛ ɾ tʰ i] * გელა: [ɡ ɛ l ɑ] * პენი: [pʼ ɛ n i] |
Occurrences: 5,587 Examples: * ნახონ: [n ɑ x ɔ n] * თორი: [tʰ ɔ ɾ i] * როცა: [ɾ ɔ ts ɑ] * ზოლს: [z ɔ l s] |
|||
Open |
Occurrences: 14,642 Examples: * თაღია: [tʰ ɑ ɣ i ɑ] * ჯამში: [dʒ ɑ m ʃ i] * ძმათა: [dz m ɑ tʰ ɑ] * ნახონ: [n ɑ x ɔ n] |