Tamil 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 tamil_cv
Or download from the release page.
Intended use#
This dictionary is intended for forced alignment of Tamil transcripts.
This dictionary uses the XPF phone set for Tamil, and was used in training the Tamil 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 |
Dental |
Alveolar |
Alveopalatal |
Retroflex |
Palatal |
Velar |
Glottal |
---|---|---|---|---|---|---|---|---|---|
Nasal |
Occurrences: 53,628 Examples: * "நமது: [n̪ a m a t̪ u] * மஸ்கா: [m a s k aː] * னிடம்: [n i ʈ a m] * சிம்ம: [t͡ʃ i m m a] |
Occurrences: 21,254 Examples: * நல்லன: [n̪ a l l a n a] * "நமது: [n̪ a m a t̪ u] * 'நியூ: [n̪ i j uː] * "இந்த: [i n̪ t̪ a] |
Occurrences: 44,762 Examples: * நல்லன: [n̪ a l l a n a] * சோழன்: [t͡ʃ a ɻ a n] * aspen: [æ s p ə n] * னிடம்: [n i ʈ a m] |
Occurrences: 13,875 Examples: * முரண்: [m u ɾ a ɳ] * உணரச்: [u ɳ a ɾ a t͡ʃ] * மணிக்: [m a ɳ i k] * மரணப்: [m a ɾ a ɳ a p] |
Occurrences: 1,432 Examples: * ஆக்ஞை: [aː k ɲ a j] * அஞ்ச: [a ɲ t͡ʃ a] * ஞாலமே: [ɲ aː l a m eː] * இஞ்சு: [i ɲ t͡ʃ u] |
Occurrences: 8,005 Examples: * ஆங்க: [aː ŋ k a] * தங்கி: [t̪ a ŋ k i] * லங்கா: [l a ŋ k aː] * hmong: [m ɔ ŋ] |
|||
Stop |
Occurrences: 64,119 Examples: * பல,: [p a l a] * பிரேத: [p i ɾ eː t̪ a] * பலவகை: [p a l a ʋ a k a j] * aspen: [æ s p ə n] Occurrences: 320 Examples: * bunny: [b ʌ n i] * bret: [b ɹ ɛ t] * boat": [b o w t] * db-ல்: [d b l] |
Occurrences: 89,927 Examples: * "நமது: [n̪ a m a t̪ u] * சக்த: [t͡ʃ a k t̪ a] * பிரேத: [p i ɾ eː t̪ a] * தள்ளு: [t̪ a ɭ ɭ u] |
Occurrences: 546 Examples: * tuen: [t w ə n] * bret: [b ɹ ɛ t] * boat": [b o w t] * kilty: [k ɪ l t i] Occurrences: 363 Examples: * db-ல்: [d b l] * duane: [d w e j n] * eddie: [ɛ d i] * hard: [h ɑ ɹ d] |
Occurrences: 56,064 Examples: * னிடம்: [n i ʈ a m] * கிட: [k i ʈ a] * கஷ்ட: [k a ʂ ʈ a] * அடடா!: [a ʈ a ʈ aː] |
Occurrences: 113,775 Examples: * அகரரை: [a k a ɾ a ɾ a j] * சக்த: [t͡ʃ a k t̪ a] * பலவகை: [p a l a ʋ a k a j] * மஸ்கா: [m a s k aː] Occurrences: 170 Examples: * big: [b ɪ ɡ] * gavin: [ɡ æ v ɪ n] * sega: [s i ɡ ə] * targa: [t ɑ ɹ ɡ ə] |
||||
Affricate |
Occurrences: 27,617 Examples: * சக்த: [t͡ʃ a k t̪ a] * சோழன்: [t͡ʃ a ɻ a n] * சிம்ம: [t͡ʃ i m m a] * விதச்: [ʋ i t̪ a t͡ʃ] Occurrences: 1,510 Examples: * கஜ்ஜை: [k a d͡ʒ d͡ʒ a j] * ஜோவை: [d͡ʒ a ʋ a j] * பூஜ்ய: [p uː d͡ʒ j a] * ஜல்: [d͡ʒ a l] |
||||||||
Sibilant |
Occurrences: 3,761 Examples: * மஸ்கா: [m a s k aː] * aspen: [æ s p ə n] * ஓக்ஸ்: [oː k s] * கம்ஸ்: [k a m s] Occurrences: 196 Examples: * zuni.: [z u n i] * bends: [b ɛ n d z] * james: [d͡ʒ e j m z] * quiz: [k w ɪ z] |
Occurrences: 96 Examples: * crash: [k ɹ æ ʃ] * busch: [b ʊ ʃ] * shire: [ʃ a j ɹ] * shea: [ʃ e j] Occurrences: 3 Examples: |
Occurrences: 1,364 Examples: * கஷ்ட: [k a ʂ ʈ a] * மேஷ: [m eː ʂ a] * ஹர்ஷ: [h a ɾ ʂ a] * சோஷல்: [t͡ʃ a ʂ a l] |
||||||
Fricative |
Occurrences: 138 Examples: * fine: [f a j n] * foloi: [f ɑ l o j] * fen: [f ɛ n] * roof: [ɹ u f] Occurrences: 119 Examples: * "love: [l ʌ v] * avi: [ɑ v i] * gavin: [ɡ æ v ɪ n] * heavy: [h ɛ v i] |
Occurrences: 53 Examples: * thumb: [θ ʌ m] * north: [n ɔ ɹ θ] * athy: [æ θ i] * south: [s a w θ] Occurrences: 14 Examples: * "the: [ð ə] * this: [ð ɪ s] * the: [ð ə] * “the: [ð ə] |
Occurrences: 973 Examples: * கோஹன்: [k a h a n] * ஹர்ஷ: [h a ɾ ʂ a] * ஹீட்: [h iː ʈ] * ஹினா: [h i n aː] |
||||||
Approximant |
Occurrences: 410 Examples: * tuen: [t w ə n] * nowy: [n a w i] * joey: [d͡ʒ o w i] * boat": [b o w t] |
Occurrences: 45,624 Examples: * பலவகை: [p a l a ʋ a k a j] * ஏவிய: [eː ʋ i j a] * வேலை,: [ʋ eː l a j] * அவதி: [a ʋ a t̪ i] |
Occurrences: 621 Examples: * bret: [b ɹ ɛ t] * porte: [p ɔ ɹ t] * hard: [h ɑ ɹ d] * crew: [k ɹ u] Occurrences: 301 Examples: * aris: [ɹ̩ i s] * aamer: [e j m ɹ̩] * early: [ɹ̩ l i] * “mr.: [m ɪ s t ɹ̩] |
Occurrences: 8,751 Examples: * சோழன்: [t͡ʃ a ɻ a n] * அழகா?: [a ɻ a k aː] * பொழு: [p a ɻ u] * உழை,: [u ɻ a j] |
Occurrences: 75,375 Examples: * அகரரை: [a k a ɾ a ɾ a j] * பலவகை: [p a l a ʋ a k a j] * ஏவிய: [eː ʋ i j a] * வேலை,: [ʋ eː l a j] |
||||
Tap |
Occurrences: 56,008 Examples: * அகரரை: [a k a ɾ a ɾ a j] * பிரேத: [p i ɾ eː t̪ a] * முரண்: [m u ɾ a ɳ] * ரூவா: [ɾ uː ʋ aː] |
||||||||
Trill |
Occurrences: 32,310 Examples: * அற்றே: [a r r eː] * பெறக்: [p e r a k] * மாறின: [m aː r i n a] * சீரற்: [t͡ʃ iː ɾ a r] |
||||||||
Lateral |
Occurrences: 41,327 Examples: * நல்லன: [n̪ a l l a n a] * பல,: [p a l a] * பலவகை: [p a l a ʋ a k a j] * வேலை,: [ʋ eː l a j] |
||||||||
Lateral Tap |
Occurrences: 30,631 Examples: * தள்ளு: [t̪ a ɭ ɭ u] * வடதளி: [ʋ a ʈ a t̪ a ɭ i] * டபுள்: [ʈ a p u ɭ] * வேளை.: [ʋ eː ɭ a j] |
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: 102,174 Examples: * பிரேத: [p i ɾ eː t̪ a] * ஏவிய: [eː ʋ i j a] * னிடம்: [n i ʈ a m] * சிம்ம: [t͡ʃ i m m a] Occurrences: 5,061 Examples: * ஈதல்: [iː t̪ a l] * கீழே: [k iː ɻ eː] * சீராக: [t͡ʃ iː ɾ aː k a] * சீனி.: [t͡ʃ iː n i] |
Occurrences: 96,143 Examples: * "நமது: [n̪ a m a t̪ u] * தள்ளு: [t̪ a ɭ ɭ u] * முரண்: [m u ɾ a ɳ] * பொழு: [p a ɻ u] Occurrences: 5,972 Examples: * 'நியூ: [n̪ i j uː] * ரூவா: [ɾ uː ʋ aː] * சூடிய: [t͡ʃ uː ʈ i j a] * சூது,: [t͡ʃ uː t̪ u] |
|||
Occurrences: 467 Examples: * kilty: [k ɪ l t i] * essex: [ɛ s ɪ k s] * big: [b ɪ ɡ] * gavin: [ɡ æ v ɪ n] |
Occurrences: 42 Examples: * woman: [w ʊ m ə n] * brook: [b ɹ ʊ k] * busch: [b ʊ ʃ] * look: [l ʊ k] |
||||
Close-Mid |
Occurrences: 16,441 Examples: * எனல்: [e n a l] * யென்ன: [j e n n a] * பெறக்: [p e r a k] * எழு.: [e ɻ u] Occurrences: 17,676 Examples: * பிரேத: [p i ɾ eː t̪ a] * ஏவிய: [eː ʋ i j a] * வேலை,: [ʋ eː l a j] * அற்றே: [a r r eː] |
Occurrences: 1,988 Examples: * ஒரியா: [o ɾ i j aː] * ஒதும்: [o t̪ u m] * ஒடின்: [o ʈ i n] * foloi: [f ɑ l o j] Occurrences: 572 Examples: * ஓக்ஸ்: [oː k s] * ஓடத்: [oː ʈ a t̪] * ஓங்கி: [oː ŋ k i] * ஓலைத்: [oː l a j t̪] |
|||
Occurrences: 780 Examples: * aspen: [æ s p ə n] * tuen: [t w ə n] * sega: [s i ɡ ə] * targa: [t ɑ ɹ ɡ ə] |
|||||
Open-Mid |
Occurrences: 333 Examples: * bret: [b ɹ ɛ t] * eddie: [ɛ d i] * del: [d ɛ l] * fen: [f ɛ n] |
Occurrences: 100 Examples: * bunny: [b ʌ n i] * "love: [l ʌ v] * chuck: [t͡ʃ ʌ k] * rough: [ɹ ʌ f] Occurrences: 178 Examples: * hmong: [m ɔ ŋ] * porte: [p ɔ ɹ t] * all: [ɔ l] * lost: [l ɔ s t] |
|||
Occurrences: 290 Examples: * aspen: [æ s p ə n] * knapp: [n æ p] * cap: [k æ p] * ann: [æ n] |
|||||
Open |
Occurrences: 245,580 Examples: * நல்லன: [n̪ a l l a n a] * அகரரை: [a k a ɾ a ɾ a j] * பல,: [p a l a] * "நமது: [n̪ a m a t̪ u] Occurrences: 65,497 Examples: * மஸ்கா: [m a s k aː] * பாபா: [p aː p aː] * அழகா?: [a ɻ a k aː] * அடடா!: [a ʈ a ʈ aː] |
Occurrences: 355 Examples: * foloi: [f ɑ l o j] * avi: [ɑ v i] * hard: [h ɑ ɹ d] * targa: [t ɑ ɹ ɡ ə] |
Diphthongs#
aʊ