English (Nigeria) MFA dictionary v2.0.0#
@techreport{mfa_english_nigeria_mfa_dictionary_2022,
author={McAuliffe, Michael and Sonderegger, Morgan},
title={English (Nigeria) MFA dictionary v2.0.0},
address={\url{https://mfa-models.readthedocs.io/pronunciation dictionary/English/English (Nigeria) MFA dictionary v2_0_0.html}},
year={2022},
month={Mar},
}
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.
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: 9,914 Examples: * mers: [m aː z] * vomer: [v o m a] * pump: [pʰ ɔ m p] * how'm: [h aw m] Occurrences: 2,445 Examples: * gammy: [ɡ a mʲ i] * amuse: [a mʲ uː z] * miz: [mʲ i z] * mewl: [mʲ uː ɫ] |
Occurrences: 570 Examples: * info: [i ɱ f o] * envy: [ɛ ɱ vʲ i] * envoy: [ɛ ɱ v ɔj] * infra: [i ɱ f ɹ a] |
Occurrences: 19,830 Examples: * oron: [ɔ ɹ ɔ n] * capon: [cʰ e p u n] * niall: [n aj a ɫ] * nuns: [n ɔ n z] |
Occurrences: 5,022 Examples: * niall: [ɲ iː ɫ] * sony: [s o ɲ i] * unix: [j uː ɲ i k s] * ani: [a ɲ i] |
Occurrences: 3,762 Examples: * tango: [tʰ a ŋ ɡ o] * franc: [f ɹ a ŋ k] * swing: [s w i ŋ ɡ] * pongy: [pʰ ɔ ŋ i] |
|||
Stop |
Occurrences: 5,581 Examples: * capon: [cʰ e p u n] * pump: [pʰ ɔ m p] * alps: [a ɫ p s] * vamp: [v a m p] Occurrences: 6,857 Examples: * batty: [b a tʰ i] * blag: [b l a ɡ] * burke: [b aː k] * bused: [b ɔ s t] |
Occurrences: 15,246 Examples: * jolt: [dʒ o ɫ t] * ate: [e t] * bused: [b ɔ s t] * bight: [b aj t] Occurrences: 10,431 Examples: * dotty: [d ɔ tʰ i] * indus: [i n d u s] * drear: [d ɹ i a] * deafs: [d ɛ f s] |
Occurrences: 2,814 Examples: * query: [c w i ɹ i] * clap: [c ʎ a p] * kukri: [kʰ ʊ c ɹ i] * clad: [c ʎ a d] Occurrences: 1,758 Examples: * ogive: [o ɟ aj v] * guild: [ɟ i ɫ d] * girl: [ɟ ɛː ɫ] * ghit: [ɟ i t] |
Occurrences: 9,658 Examples: * burke: [b aː k] * hoax: [h o k s] * coca: [kʰ o k a] * folk: [f o k] Occurrences: 5,584 Examples: * blag: [b l a ɡ] * hogo: [h o ɡ o] * gammy: [ɡ a mʲ i] * dig: [dʲ i ɡ] |
Occurrences: 35 Examples: * ate: [e ʔ] * but: [b ɔ ʔ] * and: [ʔ a n] * fight: [f aj ʔ] |
|||
Affricate |
Occurrences: 2,144 Examples: * chili: [tʃ i ʎ i] * stew: [s tʃ uː] * rutch: [ɹ ɔ tʃ] * dacha: [d a tʃ a] Occurrences: 3,468 Examples: * joys: [dʒ ɔj z] * jolt: [dʒ o ɫ t] * ogive: [o dʒ aj v] * gyre: [dʒ aj a] |
|||||||
Sibilant |
Occurrences: 22,289 Examples: * hoax: [h o k s] * suvs: [ɛ s j uː vʲ iː z] * alps: [a ɫ p s] * stew: [s tʃ uː] Occurrences: 7,465 Examples: * flues: [f l uː z] * mers: [m aː z] * joys: [dʒ ɔj z] * suvs: [ɛ s j uː vʲ iː z] |
Occurrences: 4,495 Examples: * ship: [ʃ i p] * shady: [ʃ e dʲ i] * plash: [p l a ʃ] * lycia: [ʎ i ʃ a] Occurrences: 376 Examples: * loge: [l o ʒ] * ush: [j uː ʒ] * asian: [e ʒ a n] * gigue: [ʒ iː ɡ] |
||||||
Fricative |
Occurrences: 5,227 Examples: * flues: [f l uː z] * faves: [f e v z] * three: [f ɹ iː] * folk: [f o k] Occurrences: 1,479 Examples: * phi: [fʲ iː] * fixed: [fʲ i k s t] * beefy: [bʲ iː fʲ i] * fure: [fʲ ʊ] Occurrences: 3,503 Examples: * vomer: [v o m a] * vamp: [v a m p] * faves: [f e v z] * coven: [kʰ ɔ v ɛ n] Occurrences: 1,019 Examples: * suvs: [ɛ s j uː vʲ iː z] * aaav: [e e e vʲ iː] * gravy: [ɟ ɹ e vʲ i] * davit: [d a vʲ i t] |
Occurrences: 1,720 Examples: * three: [θ ɹ iː] * thor: [θ ɔ] * theia: [θ iː a] * jth: [dʒ e θ] Occurrences: 411 Examples: * th': [ð] * paths: [pʰ a ð z] * then: [ð ɛ n] * rathe: [ɹ e ð] |
Occurrences: 647 Examples: * hail: [ç e ɫ] * ghit: [dʒ iː ç i t] * ohim: [o ç i m] * hygge: [ç uː ɡ ɛ] |
Occurrences: 2,229 Examples: * hoax: [h o k s] * how'm: [h aw m] * henry: [h ɛ n ɹ i] * hogo: [h o ɡ o] |
||||
Approximant |
Occurrences: 3,243 Examples: * query: [c w i ɹ i] * whim: [w i m] * swift: [s w i f t] * wayne: [w e n] |
Occurrences: 16,947 Examples: * oron: [ɔ ɹ ɔ n] * henry: [h ɛ n ɹ i] * three: [f ɹ iː] * prays: [p ɹ e z] |
Occurrences: 1,086 Examples: * suvs: [ɛ s j uː vʲ iː z] * yin: [j i n] * usta: [j uː s t a] * unix: [j uː ɲ i k s] |
|||||
Lateral |
Occurrences: 8,120 Examples: * flues: [f l uː z] * blag: [b l a ɡ] * sloom: [s l uː m] * lazar: [l e z a] Occurrences: 6,888 Examples: * jolt: [dʒ o ɫ t] * hail: [ç e ɫ] * niall: [n aj a ɫ] * alps: [a ɫ p s] |
Occurrences: 5,499 Examples: * chili: [tʃ i ʎ i] * ellis: [ɛ ʎ i s] * clap: [c ʎ a p] * cully: [kʰ ɔ ʎ 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: 37,694 Examples: * batty: [b a tʰ i] * henry: [h ɛ n ɹ i] * nike: [n aj cʰ i] * chili: [tʃ i ʎ i] Occurrences: 6,518 Examples: * suvs: [ɛ s j uː vʲ iː z] * niall: [ɲ iː ɫ] * chili: [tʃ i ʎ iː] * three: [f ɹ iː] |
Occurrences: 3,895 Examples: * capon: [cʰ e p u n] * indus: [i n d u s] * melon: [m ɛ l u n] * cyrus: [s aj ɹ u s] Occurrences: 4,365 Examples: * flues: [f l uː z] * suvs: [ɛ s j uː vʲ iː z] * stew: [s tʃ uː] * sloom: [s l uː m] |
|||
Occurrences: 3,482 Examples: * rutch: [ɹ ʊ tʃ] * look: [l ʊ k] * kukri: [kʰ ʊ c ɹ i] * dure: [dʲ ʊ a] |
|||||
Close-Mid |
Occurrences: 6,864 Examples: * capon: [cʰ e p u n] * hail: [ç e ɫ] * faves: [f e v z] * prays: [p ɹ e z] |
Occurrences: 6,443 Examples: * hoax: [h o k s] * jolt: [dʒ o ɫ t] * coca: [kʰ o k a] * vomer: [v o m a] |
|||
Open-Mid |
Occurrences: 16,455 Examples: * suvs: [ɛ s j uː vʲ iː z] * henry: [h ɛ n ɹ i] * ate: [ɛ t] * ellis: [ɛ ʎ i s] Occurrences: 679 Examples: * girl: [ɟ ɛː ɫ] * glare: [ɡ l ɛː] * fare: [f ɛː] * arian: [ɛː ɹ i a n] |
Occurrences: 135 Examples: * perv: [pʰ ɜ v] * myrrh: [m ɜ] * blur: [b l ɜ] * furor: [fʲ ɜ ɹ ɔ] |
Occurrences: 21,344 Examples: * jolt: [dʒ ɔ ɫ t] * oron: [ɔ ɹ ɔ n] * pump: [pʰ ɔ m p] * nuns: [n ɔ n z] |
||
Open |
Occurrences: 32,183 Examples: * batty: [b a tʰ i] * blag: [b l a ɡ] * coca: [kʰ o k a] * vomer: [v o m a] Occurrences: 1,875 Examples: * mers: [m aː z] * burke: [b aː k] * girl: [ɡ aː ɫ] * lure: [l aː] |
Diphthongs#
aj
aw
ɔj