English (Nigeria) MFA dictionary v3.0.0#
@techreport{mfa_english_nigeria_mfa_dictionary_2024,
author={McAuliffe, Michael and Sonderegger, Morgan},
title={English (Nigeria) MFA dictionary v3.0.0},
address={\url{https://mfa-models.readthedocs.io/pronunciation dictionary/English/English (Nigeria) MFA dictionary v3_0_0.html}},
year={2024},
month={Feb},
}
G2P models |
Installation#
Install from the MFA command line:
mfa model download dictionary english_nigeria_mfa
Or download from the release page.
The dictionary available from the release page and command line installation has pronunciation and silence probabilities estimated as part acoustic model training (see Silence probability format and training pronunciation probabilities for more information. If you would like to use the version of this dictionary without probabilities, please see the [plain dictionary](https://raw.githubusercontent.com/MontrealCorpusTools/mfa-models/main/dictionary/english/mfa/English (Nigeria) MFA dictionary v3_0_0.dict).
Intended use#
This dictionary is intended for forced alignment of English transcripts.
This dictionary uses the MFA phone set for English, and was used in training the English MFA 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 |
Palatal |
Velar |
Glottal |
---|---|---|---|---|---|---|---|---|
Nasal |
Occurrences: 11,447 Examples: * maid: [m e d] * mario: [m a ɹ i o] * maps: [m a p s] * email: [ɛ m e ɫ] Occurrences: 2,723 Examples: * mist: [mʲ i s t] * admit: [a d mʲ i t] * gimme: [ɟ i mʲ iː] * myth: [mʲ i θ] |
Occurrences: 11 Examples: |
Occurrences: 23,049 Examples: * agent: [e dʒ ɛ n t] * noise: [n ɔj z] * wound: [w uː n d] * down: [d aw n] |
Occurrences: 4,919 Examples: * until: [ɔ ɲ tʲ i ɫ] * danny: [d a ɲ i] * minim: [mʲ i ɲ i m] * any: [ɛ ɲ i] |
Occurrences: 4,926 Examples: * lying: [l aj i ŋ] * hang: [h a ŋ] * bling: [b ʎ i ŋ ɡ] * mango: [m a ŋ ɡ o] |
|||
Stop Plain |
Occurrences: 6,277 Examples: * april: [e p ɹ i ɫ] * plato: [p l e tʰ o] * prove: [p ɹ o v] * spite: [s p aj] Occurrences: 1,118 Examples: * pure: [pʲ ɔ] * lapis: [l a pʲ i s] * april: [e pʲ i ɫ] * puree: [pʲ ʊ ɹ iː] Occurrences: 9 Examples: Occurrences: 7,632 Examples: * rib: [ɹ i b] * bow: [b o] * biro: [b aj ɹ o] * beige: [b e ʃ] Occurrences: 1,475 Examples: * bids: [bʲ i d z] * obeah: [o bʲ i a] * abc: [e bʲ iː s iː] * bida: [bʲ i d a] |
Occurrences: 93 Examples: * thou: [t̪ aw] * death: [d ɛ t̪] * thing: [t̪ i ŋ] * third: [t̪ aː d] Occurrences: 70 Examples: * these: [d̪ iː s] * other: [ɔ d̪ a] * their: [d̪ ɛː] * mouth: [m aw d̪] |
Occurrences: 16,736 Examples: * stain: [s t e n] * treat: [t ɹ iː t] * split: [s p ʎ i t] * least: [ʎ iː s t] Occurrences: 4,694 Examples: * unity: [j uː ɲ i tʲ i] * pity: [pʰ i tʲ i] * tune: [tʲ uː n] * multi: [m ɔ ɫ tʲ i] Occurrences: 95 Examples: * twin: [tʷ i n] * twill: [tʷ i ɫ] * twist: [tʷ i s t] * twins: [tʷ i n z] Occurrences: 12,967 Examples: * dance: [d a n s] * card: [kʰ a d] * you'd: [j uː d] * okada: [o kʰ a d a] Occurrences: 3,352 Examples: * india: [i ɲ dʲ i a] * diya: [dʲ iː a] * dues: [dʲ uː z] * deans: [dʲ iː n z] |
Occurrences: 2,528 Examples: * naked: [n e c i d] * clips: [c ʎ i p s] * skies: [s c aj z] * scars: [s c aː z] Occurrences: 304 Examples: * queen: [cʷ iː n] * quill: [cʷ i ɫ] * query: [cʷ i ɹ i] * quit: [cʷ i t] Occurrences: 1,047 Examples: * giver: [ɟ i v a] * green: [ɟ ɹ iː n] * gives: [ɟ i v s] * agree: [a ɟ ɹ iː] Occurrences: 64 Examples: |
Occurrences: 10,174 Examples: * think: [θ i ŋ k] * clef: [k l ɛ f] * score: [s k ɔ] * dock: [d ɔ k] Occurrences: 473 Examples: * choir: [kʷ aj a] * akwa: [a kʷ a] * quake: [kʷ e k] * squad: [s kʷ ɔ d] Occurrences: 7,731 Examples: * gad: [ɡ a d] * egos: [ɛ ɡ o z] * leg: [l ɛ ɡ] * rogue: [ɹ o ɡ] Occurrences: 45 Examples: * gweje: [ɡʷ ɛ dʒ e] |
Occurrences: 13 Examples: * that: [d̪ a ʔ] * put: [pʰ ʊ ʔ] * jet: [dʒ ɛ ʔ] * seat: [s iː ʔ] |
||
Aspirated |
Occurrences: 4,634 Examples: * pence: [pʰ ɛ n s] * paves: [pʰ e v z] * pens: [pʰ ɛ n z] * polka: [pʰ ɔ ɫ k a] |
Occurrences: 4,860 Examples: * tare: [tʰ ɛː] * two: [tʰ uː] * team: [tʰ iː m] * too: [tʰ uː] |
Occurrences: 1,654 Examples: * kills: [cʰ i ɫ z] * nyaki: [ɲ a cʰ i] * kebbi: [cʰ ɛ bʲ i] * kemi: [cʰ ɛ mʲ i] |
Occurrences: 5,045 Examples: * cause: [kʰ ɔ s] * cola: [kʰ o l a] * kandy: [kʰ a ɲ dʲ i] * calf: [kʰ a f] |
||||
Affricate |
Occurrences: 2,424 Examples: * fetch: [f ɛ tʃ] * rich: [ɹ i tʃ] * truth: [tʃ uː t̪] * touch: [tʰ ɔ tʃ] Occurrences: 3,814 Examples: * badge: [b a dʒ] * jobs: [dʒ ɔ p s] * john: [dʒ ɔ n] * jar: [dʒ a] |
|||||||
Sibilant |
Occurrences: 25,322 Examples: * slip: [s ʎ i p] * scoop: [s k uː p] * serum: [s iː u m] * asita: [a s i t a] Occurrences: 9,061 Examples: * polls: [pʰ o ɫ z] * pens: [pʰ ɛ n z] * kinds: [kʰ aj n z] * zeal: [z iː ɫ] |
Occurrences: 5,476 Examples: * shey: [ʃ e] * shove: [ʃ ɔ v] * clash: [k l a ʃ] * wash: [w ɔ ʃ] |
||||||
Fricative |
Occurrences: 5,765 Examples: * fault: [f ɔ ɫ t] * fight: [f aj t] * firs: [f aː s] * fold: [f o ɫ d] Occurrences: 1,608 Examples: * fuel: [fʲ ʊ ɛ ɫ] * fifty: [fʲ i f tʰ i] * fear: [fʲ i a] * few: [fʲ uː] Occurrences: 3,935 Examples: * move: [m uː v] * via: [v aj a] * river: [ɹ aj v a] * delve: [d ɛ ɫ v] Occurrences: 1,126 Examples: * davis: [d e vʲ i s] * vicky: [vʲ i cʰ i] * views: [vʲ uː s] * vera: [vʲ i ɹ a] |
Occurrences: 1,798 Examples: * worth: [w ɔ θ] * sith: [s iː θ] * maths: [m a θ s] * theft: [θ ɛ f t] Occurrences: 441 Examples: * oaths: [o ð z] * dey: [ð e] * other: [ɔ ð a] * these: [ð iː z] |
Occurrences: 694 Examples: * heath: [ç iː θ] * heavy: [ç iː vʲ i] * huge: [ç uː dʒ] * haze: [ç e z] |
Occurrences: 2,422 Examples: * hand: [h a n d] * hole: [h o ɫ] * hire: [h aj a] * hans: [h a n s] |
||||
Approximant |
Occurrences: 2,589 Examples: * white: [w aj] * wider: [w aj d a] * wild: [w aj ɫ d] * ewedu: [ɛ w ɛ d uː] |
Occurrences: 19,115 Examples: * trim: [tʃ ɹ i m] * risk: [ɹ i s k] * guru: [ɡ uː ɹ uː] * react: [ɹ iː a t] |
Occurrences: 1,209 Examples: * yaba: [j a b a] * yale: [j e ɫ] * azure: [a z j ʊ a] * ijaye: [iː dʒ a j e] |
|||||
Lateral |
Occurrences: 9,196 Examples: * flame: [f l e m] * late: [l e t] * clerk: [k l a k] * los: [l ɔ s] Occurrences: 7,411 Examples: * hull: [h ɔ ɫ] * cells: [s ɛ ɫ s] * meals: [mʲ iː ɫ z] * fails: [f e ɫ z] |
Occurrences: 5,843 Examples: * badly: [b a d ʎ i] * folic: [f o ʎ i k] * early: [aː ʎ i] * sleep: [s ʎ iː p] |
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: 40,457 Examples: * mints: [mʲ i n t s] * being: [bʲ iː i ŋ] * tenth: [tʰ i n θ] * bobby: [b ɔ bʲ i] Occurrences: 6,260 Examples: * meets: [mʲ iː t s] * deity: [dʲ iː i tʲ i] * read: [ɹ iː d] * leans: [ʎ iː n z] |
Occurrences: 4,063 Examples: * rome: [ɹ u m] * torus: [tʰ ɔ ɹ u s] * yusuf: [j u s ʊ f] * mosun: [m ʊ s u n] Occurrences: 4,739 Examples: * fruit: [f ɹ uː t] * who's: [h uː z] * tool: [tʃ uː ɫ] * umar: [uː m a ɹ] |
|||
Occurrences: 3,803 Examples: * puts: [pʰ ʊ t s] * pulls: [pʰ ʊ ɫ z] * moot: [m ʊ t] * old: [ɔ ʊ ɫ d] |
|||||
Close-Mid |
Occurrences: 7,803 Examples: * flame: [f l e m] * aim: [e m] * saint: [s e n t] * wade: [w e d] |
Occurrences: 7,104 Examples: * don't: [d o n] * robe: [ɹ o b] * oyo: [ɔj o] * nodes: [n o t s] |
|||
Open-Mid |
Occurrences: 21,132 Examples: * pedal: [pʰ ɛ d a ɫ] * tense: [tʰ ɛ n s] * jetty: [dʒ ɛ tʰ i] * trend: [tʃ ɹ ɛ n] Occurrences: 769 Examples: * girl: [ɡ ɛː ɫ] * chair: [tʃ ɛː] * are: [ɛː] * bear: [b ɛː] |
Occurrences: 136 Examples: * bury: [b ɜ ɹ i] Occurrences: 3 Examples: |
Occurrences: 24,024 Examples: * honey: [h ɔ ɲ i] * lung: [l ɔ ŋ ɡ] * more: [m ɔ] * false: [f ɔ ɫ s] |
||
Open |
Occurrences: 35,564 Examples: * basal: [b a s a ɫ] * blast: [b l a s t] * tower: [tʰ aw a] * buyer: [b aj a] Occurrences: 2,169 Examples: * versa: [v aː s a] * turn: [tʰ aː n] * birth: [b aː s] * nerve: [n aː v] |
Diphthongs#
aj
aw
ɔj