English Language
Download
This chapter describes the linguistic resources included in the file
eng.zip of the lang folder.
List of phonemes
The following list of phonemes includes both the British English and the American English. The acoustic model contains all of them. However, only the American English pronunciation dictionary is provided. The given examples are for American English.
SPPAS is based on the X-SAMPA standard. See https://en.wikipedia.org/wiki/X-SAMPA for details.
Consonant Plosives
| SPPAS | IPA | Description | Examples |
|---|---|---|---|
| p | p | voiceless bilabial | pie, spy, cap |
| b | b | voiced bilabial | buy, cab |
| t | t | voiceless alveolar | tie, sty, cat, atom |
| d | d | voiced alveolar | dye, cad, do |
| k | k | voiceless velar | sky, crack, quick |
| g | g | voiced velar | guy, bag, luggage |
Consonant Fricatives
| SPPAS | IPA | Description | Examples |
|---|---|---|---|
| D | ð | voiced dental | thy, breathe, father |
| f | f | voiceless labiodental | phi, caff, fan |
| s | s | voiceless alveolar | sigh, mass |
| S | ʃ | voiceless postalveolar | shy, cash, emotion |
| z | z | voiced alveolar | zoo, has |
| Z | ʒ | voiced postalveolar | equation, pleasure, vision, beige |
| v | v | voiced labiodental | vie, have |
| T | θ | voiceless dental | thigh, math |
| h | h | voiceless glottal | high, ahead |
Consonant Nasals
| SPPAS | IPA | Description | Examples |
|---|---|---|---|
| m | m | bilabial | my, smile, cam |
| n | n | alveolar | nigh, snide, can |
| N | ŋ | voiced velar | sang, sink, singer |
Consonant Liquids
| SPPAS | IPA | Description | Examples |
|---|---|---|---|
| l | l | alveolar lateral | lie, sly, gal |
| 4 | ɾ | alveolar flap | lyda, maddy, makita |
| r\ | ɹ | alveolar approximant | red, try, very |
Semivowels
| SPPAS | IPA | Description | Examples |
|---|---|---|---|
| j | j | voiced palatal | yes, yacht, william |
| w | w | voiced labiovelar | wye, swine, why |
Vowels
| SPPAS | IPA | Description | Examples |
|---|---|---|---|
| E | ɛ | open-mid front unrounded | dress, bed, fell, men |
| A: | ɑ: | open back unrounded | palm, father, bra |
| A | ɒ | open back rounded | lot, pod, John |
| O: | ɔ: | open-mid back rounded | thought, Maud, dawn, fall |
| V | ʌ | open-mid back unrounded | strut, mud, dull, gun |
| i | i | close front unrounded | happy, serious |
| i: | i: | close front unrounded | fleece, seed, feel, sea |
| u: | u: | close back rounded | goose, food, chew, do |
| @ | ə | schwa | a, baccus |
| I | ɪ | near-close near-front unrounded | kit, lid, fill, bin |
| u | u | close back rounded vowel | absolute, assume |
| U | ʊ | near-close near-back rounded | foot, full, woman |
| { | æ | near-open front unrounded | trap, pad, shall, ban |
Affricates
| SPPAS | IPA | Description | Examples |
|---|---|---|---|
| dZ | d͡ʒ | voiced postalveolar | giant, badge, jam |
| tS | t͡ʃ | voiceless postalveolar | China, catch |
Other symbols
| SPPAS | IPA | Examples |
|---|---|---|
| aI | aɪ | price, ride, file, pie |
| aU | aʊ | mouth, loud, down, how |
| eI | eɪ | face, fail, vein, pay |
| OI | ɔɪ | choice, void, foil, boy |
| @U | oʊ | goat, code, foal, go |
| 3:r | ɜ:r | liner, foundered, current |
Recently added (not documented yet)
| SPPAS | IPA | Description |
|---|---|---|
| @U | ||
| E@ | ||
| 3: | ɜ | open-mid central unrounded vowel |
| I@ | ||
| l= | /l/ syllabic | |
| n= | /n/ syllabic | |
| Q | ɒ | open back rounded vowel |
Fillers
| SPPAS | Description |
|---|---|
| laugh | laughter |
| noise | noises, unintelligible speech |
| dummy | un-transcribed speech |
Lexicons
All English lexicons are (c) CNRS, Laboratoire Parole et Langage, Aix-en-Provence, France:
eng.vocabcontains a list of 120k different words;eng.stpis a list of 150 stop-words;eng_num.replallows to convert numbers to their written form;eng.replallows to convert symbols and abbreviations into a text form.
All of them are distributed under the terms of the GNU General Public License v3.
Pronunciation dictionary
The pronunciation dictionary is for North American English. It was downloaded in 2011 from the CMU web page. This Carnegie Mellon Pronouncing Dictionary (version 0.6) is Copyright (C) 1993-2008 by Carnegie Mellon University. We acknowledge CMU for distributing freely this resource and allowing its re-distribution.
Brigitte Bigi converted the original CMUdict encoded with ARPAbet into X-SAMPA and converted the format of the file in HTK-ASCII.
Acoustic Model
The first version of the acoustic model was context-dependent (better
accuracy) but did not contain the fillers. This model distributed in
SPPAS resources was downloaded in 2014 from the VoxForge project at http://www.voxforge.org/. For the second model, the
monophones were extracted to create a new context-independent model in
which the fillers (i.e. laugh, noise and dummy) were added. The model
was under the terms of the GNU Public License
.
In 2022, a new context-independent acoustic model was trained and phonemes for British English are introduced. This model is under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International Public License
Syllabification
There's no such available resource because English syllabification can't be solved with the implemented algorithm.
Cued Speech
There is no available resource yet, but American English will be supported by mid-2025. So... check back soon!