Portuguese MFA dictionary v2.0.0#

  • Maintainer: Montreal Forced Aligner

  • Language: Portuguese

  • Dialect: N/A

  • Phone set: MFA

  • Number of words: 94,655

  • Phones: a b c d e f i j k l m n o p s t u v w x z ð ɐ ɐ̃ ɔ ɛ ɟ ɡ ɣ ɨ ɲ ɾ ʁ ʃ ʎ ʒ β

  • License: CC BY 4.0

  • Compatible MFA version: v2.0.0

  • Citation:

@techreport{mfa_portuguese_mfa_dictionary_2022,
	author={McAuliffe, Michael and Sonderegger, Morgan},
	title={Portuguese MFA dictionary v2.0.0},
	address={\url{https://mfa-models.readthedocs.io/pronunciation dictionary/Portuguese/Portuguese MFA dictionary v2_0_0.html}},
	year={2022},
	month={Mar},
}
../../_images/full_logo_yellow.svg

Installation#

Install from the MFA command line:

mfa model download dictionary portuguese_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 Portuguese transcripts.

This dictionary uses the MFA phone set for Portuguese, and was used in training the Portuguese 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

Uvular

Nasal

Occurrences:
25,982
Examples:
* amera:
[a m ɛ ɾ a]
* morsi:
[m ɔ ɾ s i]
* íome:
[i ɔ m ɨ]
* móda:
[m ɔ ð ɐ]
Occurrences:
16,904
Examples:
* naiá:
[n a ɐ j a]
* notar:
[n o t a x]
* arnez:
[ɐ ɾ n ɛ ʃ]
* hinos:
[i n u ʃ]
Occurrences:
4,338
Examples:
* alane:
[a l ɐ ɲ i]
* junho:
[ʒ u ɲ u]
* sahe:
[s ɐ ɲ ɨ]
* nhosa:
[ɲ o z ɐ]

Stop

Occurrences:
25,620
Examples:
* ipaba:
[i p a b ɐ]
* péon:
[p ɛ ]
* peito:
[p e j t u]
* prats:
[p ɾ a t s]
Occurrences:
10,233
Examples:
* ipaba:
[i p a b ɐ]
* brôôô:
[b ɾ o o o]
* broa:
[b ɾ o a]
* obtê:
[o b t e]
Occurrences:
45,991
Examples:
* retê:
[ʁ ɨ t e]
* youth:
[a j u t ɣ a]
* peito:
[p e j t u]
* prats:
[p ɾ a t s]
Occurrences:
24,754
Examples:
* fodão:
[f ɔ d ɐ̃ ]
* danse:
[d ɐ̃ s i]
* dasa:
[d a z ɐ]
* duras:
[d u ɾ ɐ s]
Occurrences:
2,452
Examples:
* raque:
[x a c i]
* kebir:
[c i b i x]
* gecy:
[ʒ e c i]
* etc:
[i t c i]
Occurrences:
663
Examples:
* segue:
[s ɛ ɟ i]
* fgts:
[ɛ f i ɟ e t e ɛ s i]
* cgt:
[s e ɟ e t e]
* guiei:
[ɟ i e j]
Occurrences:
31,769
Examples:
* quaeí:
[k w ɐ ɛ i]
* raque:
[ʁ a k ɨ]
* toca:
[t ɔ k ɐ]
* capuz:
[k a p u j s]
Occurrences:
7,022
Examples:
* gozo:
[ɡ o z u]
* gongo:
[ɡ ɡ u]
* segue:
[s ɛ ɡ ɨ]
* grego:
[ɡ ɾ e ɡ u]

Affricate

Occurrences:
4,709
Examples:
* prats:
[p ʁ a ]
* cuti:
[k u i]
* katy:
[k a i]
* frete:
[f ɾ ɛ i]
Occurrences:
2,932
Examples:
* adad:
[a d a ]
* dje:
[ ɨ]
* digam:
[ i ɡ ɐ̃ ]
* dead:
[d ɛ ]

Sibilant

Occurrences:
44,554
Examples:
* sum:
[s ]
* prats:
[p ɾ a t s]
* capuz:
[k a p u j s]
* seltz:
[s ɛ l t ʃ]
Occurrences:
12,650
Examples:
* raísa:
[x a i z ɐ]
* gozo:
[ɡ o z u]
* dasa:
[d a z ɐ]
* visar:
[v i z a x]
Occurrences:
37,771
Examples:
* capuz:
[k ɐ p u ʃ]
* seltz:
[s ɛ l t ʃ]
* jhes:
[ʒ ɨ ʃ]
* hades:
[a ð ɨ ʃ]
Occurrences:
8,260
Examples:
* joca:
[ʒ ɔ k ɐ]
* jhes:
[ʒ ɨ ʃ]
* juiz:
[ʒ u i s]
* junho:
[ʒ u ɲ u]

Fricative

Occurrences:
13,084
Examples:
* fodão:
[f ɔ d ɐ̃ ]
* aferi:
[ɐ f ɨ ɾ i]
* safou:
[s a f o w]
* vofla:
[v ɔ f l ɐ]
Occurrences:
17,612
Examples:
* visar:
[v i z a x]
* tvor:
[t v o ɾ]
* herva:
[ɛ ɾ v ɐ]
* van:
[v ɐ̃]
Occurrences:
16,618
Examples:
* odo:
[o ð u]
* ardem:
[a ɾ ð ɐ̃ ]
* hades:
[a ð ɨ ʃ]
* móda:
[m ɔ ð ɐ]
Occurrences:
9,875
Examples:
* retê:
[ʁ ɨ t e]
* raísa:
[ʁ ɐ i z ɐ]
* raque:
[ʁ a k ɨ]
* rírao:
[ʁ i ɾ a w]

Approximant

Occurrences:
12,879
Examples:
* youth:
[o w t]
* quaeí:
[k w ɐ ɛ i]
* rírao:
[ʁ i ɾ a w]
* broa:
[b ɾ o w ɐ]
Occurrences:
8,511
Examples:
* fodão:
[f ɔ d ɐ̃ ]
* orçam:
[u ɾ s ɐ̃ ]
* metam:
[m ɨ t ɐ̃ ]
* rodam:
[ʁ ɔ d ɐ̃ ]
Occurrences:
21,694
Examples:
* youth:
[a j u t ɣ a]
* naiá:
[n a ɐ j a]
* peito:
[p e j t u]
* capuz:
[k a p u j s]
Occurrences:
4,968
Examples:
* ardem:
[a ɾ ð ɐ̃ ]
* junho:
[ʒ u]
* essem:
[ɛ s ɐ̃ ]
* dalém:
[d a l ɐ̃ ]

Tap

Occurrences:
60,106
Examples:
* ardem:
[a ɾ ð ɐ̃ ]
* rírao:
[ʁ i ɾ a w]
* brôôô:
[b ɾ o o o]
* amera:
[a m ɛ ɾ a]

Lateral

Occurrences:
27,431
Examples:
* ali:
[ɐ l i]
* seltz:
[s ɛ l t ʃ]
* rolar:
[x o l a x]
* alane:
[a l ɐ ɲ i]
Occurrences:
4,826
Examples:
* ali:
[a ʎ i]
* lhos:
[ʎ u ʃ]
* bulha:
[b u ʎ ɐ]
* selig:
[s e ʎ i ɡ]

Vowels#

Vowel symbols to the left of are unrounded and those to the right are rounded.

Oral Vowels#

Front

Near-Front

Central

Near-Back

Back

Close

Occurrences:
72,602
Examples:
* ipaba:
[i p a b ɐ]
* ali:
[a ʎ i]
* quaeí:
[k w ɐ ɛ i]
* raísa:
[x a i z ɐ]
Occurrences:
34,357
Examples:
* retê:
[ʁ ɨ t e]
* raque:
[ʁ a k ɨ]
* uarte:
[w a ɾ t ɨ]
* jhes:
[ʒ ɨ ʃ]
Occurrences:
63,716
Examples:
* odo:
[o ð u]
* youth:
[a j u t ɣ a]
* peito:
[p e j t u]
* capuz:
[k a p u j s]

Close-Mid

Occurrences:
24,168
Examples:
* retê:
[ʁ ɨ t e]
* peito:
[p e j t u]
* obtê:
[o b t e]
* cachê:
[k a ʃ e]
Occurrences:
20,166
Examples:
* odo:
[o ð u]
* youth:
[o w t]
* brôôô:
[b ɾ o o o]
* broa:
[b ɾ o a]

Open-Mid

Occurrences:
11,899
Examples:
* péon:
[p ɛ ]
* quaeí:
[k w ɐ ɛ i]
* amera:
[a m ɛ ɾ a]
* seltz:
[s ɛ l t ʃ]
Occurrences:
8,975
Examples:
* toca:
[t ɔ k ɐ]
* fodão:
[f ɔ d ɐ̃ ]
* obtê:
[ɔ b t e]
* joca:
[ʒ ɔ k ɐ]
Occurrences:
71,982
Examples:
* naiá:
[n a ɐ j a]
* ipaba:
[i p a b ɐ]
* ali:
[ɐ l i]
* quaeí:
[k w ɐ ɛ i]

Open

Occurrences:
60,277
Examples:
* ardem:
[a ɾ ð ɐ̃ ]
* youth:
[a j u t ɣ a]
* naiá:
[n a ɐ j a]
* ipaba:
[i p a b ɐ]

Nasal Vowels#

Front

Near-Front

Central

Near-Back

Back

Close

Occurrences:
8,165
Examples:
* inspn:
[ s p]
* incel:
[ s ɛ w]
* minho:
[m u]
* impus:
[ p u ʃ]
Occurrences:
1,754
Examples:
* sum:
[s ]
* zumba:
[z b ɐ]
* junho:
[ʒ u]
* fungo:
[f ɡ u]

Close-Mid

Occurrences:
14,559
Examples:
* vencê:
[v s e]
* lente:
[l t ɨ]
* tente:
[t t ɨ]
* homem:
[o m ]
Occurrences:
8,436
Examples:
* péon:
[p ɛ ]
* gongo:
[ɡ ɡ u]
* rompe:
[x p]
* trons:
[t ɾ ʃ]

Open-Mid

Occurrences:
21,814
Examples:
* ardem:
[a ɾ ð ɐ̃ ]
* fodão:
[f ɔ d ɐ̃ ]
* danse:
[d ɐ̃ s i]
* orçam:
[u ɾ s ɐ̃ ]

Open