Portuguese Language

Download

This chapter describes the linguistic resources included in the file por.zip of the lang folder.

List of phonemes

Consonant Plosives

SPPAS IPA Description Examples
p p voiceless bilabial pacto
b b voiced bilabial bato
t t voiceless alveolar tacto
d d voiced alveolar dato
k k voiceless velar cacto
g g voiced velar gato

Consonant Fricatives

SPPAS IPA Description Examples
f f voiceless labiodental facto
s s voiceless alveolar saca
S ʃ voiceless postalveolar chato
z z voiced alveolar zaca
Z ʒ voiced postalveolar jacto
v v voiced labiodental vaca
x x voiceless velar rabão

Consonant Nasals

SPPAS IPA Description Examples
m m bilabial mato
n n alveolar nato
N ŋ voiced velar hong-kong
J ɲ palatal pinha

Consonant Liquids

SPPAS IPA Description Examples
l l alveolar lateral galo
L ʎ palatal lateral galho
r r alveolar trill pira
R ʁ voiced uvular rato

Semivowels

SPPAS IPA Description Examples
j j voiced palatal yoga
w w voiced labiovelar uísque

Vowels

SPPAS IPA Description Examples
E ɛ open-mid front unrounded ego, eira
a a open front unrounded parto
O ɔ open-mid back rounded pôde
i i close front unrounded hidra
e e close-mid front unrounded pega, elo
o o close-mid back rounded bola
u u close back rounded hotel
y y close front rounded emile
I ɪ near-close near-front unrounded dois
U ʊ near-close near-back rounded ido

Nasal vowels

SPPAS IPA Examples
a~ ɑ̃ anis
u~ humberto, unha

Fillers

SPPAS Description
laugh laughter
noise noises, unintelligible speech
dummy un-transcribed speech

Lexicons

All Portuguese lexicons are (c) Laboratoire Parole et Langage, Aix-en-Provence, France:

  • por.vocab contains a list of 414k different words;
  • por_num.repl allows to convert numbers to their written form;
  • por.repl allows to convert symbols and abbreviations into a text form.

All of them are distributed under the terms of the GNU General Public License.

Pronunciation Dictionary

The Portuguese pronunciation dictionary was downloaded from the Ralf catalog of dictionaries for the Simon ASR system at http://spirit.blau.in/simon/import-pls-dictionary/. It was then converted (format and phoneset) and corrected by Brigitte Bigi.

It is re-distributed under the terms of the GNU General Public License.

Acoustic Model

The acoustic model was NOT trained from data. Monophones of other models were cut and pasted to create this one, mainly from the Spanish and the French models. The Portuguese model is distributed under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International Public License.

New Portuguese data is welcome! Because data implies a better acoustic model then better alignments…