Bengali Language
Download
This chapter describes the linguistic resources included in the file
ben.zip
of the lang
folder.
The iso-639-3 code of Bengali language is BEN.
List of phonemes
Consonant Plosives
SPPAS | IPA | Description | Examples |
---|---|---|---|
b | b | voiced bilabial | |
b_h | bʰ | voiced bilabial aspirated | |
c | c | voiceless palatal | |
c_h | cʰ | voiceless palatal aspirated | |
d | d | voiced alveolar | |
d_h | dʰ | voiced alveolar aspirated | |
d| ɖ | voiced retroflex | | | d _h |
ɖʰ | voiced retroflex aspirated | |
g | g | voiced velar | |
g_h | gʰ | voiced velar | |
J\ | ɟ | voiced palatal | |
J_h | ɟʰ | voiced palatal aspirated | |
k | k | voiceless velar | |
k_h | kʰ | voiceless velar aspirated | |
p | p | voiceless bilabial | |
p_h | pʰ | voiceless bilabial aspirated | |
t | t | voiceless alveolar | |
t_h | tʰ | voiceless alveolar aspirated | |
t` | ʈ | voiceless retroflex | |
t`_h | ʈʰ | voiceless retroflex aspirated |
Notice that J and J_h were both in the first version of the pronunciation dictionary but are no longer in the current version. They remain in the acoustic model, so they can be used for Phonetization.
Affricates
SPPAS | IPA | Description | Examples |
---|---|---|---|
dZ | d͡ʒ | voiced postalveolar | |
dZ_h | d͡ʒʰ | voiced postalveolar aspirated |
Consonant Fricatives
SPPAS | IPA | Description | Examples |
---|---|---|---|
f | f | voiceless labiodental | |
h | h | voiceless glottal | |
s | s | voiceless alveolar | |
S | ʃ | voiceless postalveolar | |
v | v | voiced labiodental | |
z | z | voiced alveolar | |
Z | ʒ | voiced postalveolar |
Consonant Nasals
SPPAS | IPA | Description | Examples |
---|---|---|---|
m | m | bilabial | |
n | n | alveolar | |
N | ŋ | voiced velar |
Consonant Liquids
SPPAS | IPA | Description | Examples |
---|---|---|---|
l | l | alveolar lateral | |
r | r | alveolar trill | |
r| ɽ | voiced retroflex flap | | | r _h |
ɽʰ | voiced retroflex flap aspirated |
Semivowels
SPPAS | IPA | Description | Examples |
---|---|---|---|
j | j | voiced palatal | |
w | w | voiced labiovelar |
Vowels
SPPAS | IPA | Description | Examples |
---|---|---|---|
@ | ə | schwa | |
a | a | open front unrounded | |
{ | æ | near-open front unrounded vowel | |
e | e | close-mid front unrounded | |
i | i | close front unrounded | |
O | ɔ | open-mid back rounded | |
o | o | close-mid back rounded | |
u | u | close back rounded |
Nasal vowels (~)
SPPAS | IPA | Description | Examples |
---|---|---|---|
a~ | ã | open front unrounded nasal vowel | |
e~ | ẽ | close-mid front unrounded nasal vowel | |
i~ | ĩ | close front unrounded nasal vowel | |
O~ | ɔ̃ | open-mid back unrounded nasal vowel | |
o~ | õ | close-mid back unrounded nasal vowel | |
u~ | ũ | close back unrounded nasal vowel |
Fillers
SPPAS | Description |
---|---|
laugh | laughter |
noise | noises, unintelligible speech |
dummy | un-transcribed speech |
fp | filled pause |
Pronunciation Dictionary
The pronunciation dictionary is Copyright 2015, 2016 Google Inc. All Rights Reserved., with a CC-4.0 license. It was downloaded in October 2021, from: https://github.com/google/language-resources/tree/master/bn/data/
The phonemes have been converted to X-SAMPA and the file format to HTK-ASCII by Brigitte Bigi. Pronunciations were revised by Moumita PAKRASHI of Centre for Linguistic Science and Technology, Indian Institute of Technology Guwahati.
The dictionary is re-distributed under the terms of its original Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International Public License.
Acoustic Model
The model was developed using a Python script available in the SPPAS package:
acmtrain.py
.
This is the second version of the acoustic model. It was trained with a set of 6 manually time-aligned files (totalling about 18 seconds of speech) and 1,300 orthographically transcribed files (totalling 36 minutes of speech).
The model is distributed under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International Public License.