Regulus

Regulus is an elegantly-built procedural constructed language. Its lexicon is formed entirely from processing English words through a conversion algorithm, without exception. This concept is inspired by the 'Al Bhed' language from Final Fantasy X, which follows the same concept while using a much simpler algorithm (with far inferior results).

Word Conversion
Word conversion follows an in-depth procedure as described below, returning a converted word from the input English. First, it's determined if the word should be transliterated or translated (see below for methods). Proper names, loanwords, and the like should always be transliterated, while all other words should be translated; both methods result in new words using English letters, which are then transcribed into Regulus per the orthography tables below.

Consonants
¹Voiced variants /b/, /d/, /g/, /v/, and /z/ of voiceless consonants /p/, /t/, /k/, /f/, and /s/ are not officially included in the native alphabet—the alphabet is extended by including variants of these with an inverted breve above, using the diacritic to indicate added voice. This inverted breve above may not render properly in the table.

²Consonants /θ/, /ð/, and /w/ are not officially included in the native alphabet—the alphabet is extended by including 'þ' for /θ/ and /ð/, and 'ƿ' for /w/. Additionally, not-included /ʒ/ can be transliterated with 'w', as /ʃ/ is natively.

Transliteration
A simple transliteration routine is implemented for proper names and loanwords. Transliteration is literally letter-for-letter (accents and all), using a new alphabet, with some modifications. Note that based on phonology and orthography, it's best to replace 'x' with 'ks', 'q' with 'k', 'c' with 'k' or 's' as appropriate, and 'j' with 'dsh' prior to transliteration. As well, 'y' should be replaced with 'i', as the transcription for 'y' uses a completely different phoneme in Regulus. •'w' is transliterated as 'ƿ' •'th' is transliterated as 'þ' •'sh' is transliterated as 'w

•All other letters are transliterated literally

Translation
Translation follows the following algorithm:

1. Handle 'I' special case •If the word is the first-person pronoun 'I', return 'O' (skip remaining steps)

2. Note the casing •Denote whether the original word's casing is UPPER/lower/Proper/oTHeR

3. Screen accented & doubled characters •Replace accented characters with non-accented versions (é→e, etc.) •Replace doubled characters with singular versions (Æ→Ae, etc.)

4. Homogenize casing •Replace all upper-case characters with lower-case variants (A→a, etc.)

5. Perform common syllable replacements • Replace syllables with coded indices : ment→[000]  on→[022]    le→[041]    bi→[061] ther→[001]  se→[023]    as→[042]    by→[061] ion→[002]   ce→[023]    vi→[043]    po→[062] ing→[003]   ve→[024]    vy→[043]    tu→[063] and→[004]   si→[025]    ta→[044]    co→[064] the→[005]   sy→[025]    na→[045]    ko→[064] for→[006]   ci→[025]    ni→[046]    es→[065] pro→[007]   cy→[025]    ny→[046]    ex→[066] con→[008]   me→[026]    mi→[047]    lo→[067] com→[009]   de→[027]    my→[047]    do→[068] ver→[010]   te→[028]    fi→[048]    he→[069] cal→[011]   ne→[029]    fy→[048]    en→[070] pre→[012]   be→[030]    ca→[049]    va→[071] per→[013]   di→[031]    ka→[049]    you→[072] ble→[014]   dy→[031]    ra→[050] li→[015]    is→[032]    er→[051] ly→[015]    ys→[032]    an→[052] re→[016]    so→[033]    no→[053] ri→[017]    ma→[034]    la→[054] ry→[017]    ge→[035]    or→[055] ti→[018]    al→[036]    ha→[056] ty→[018]    we→[037]    pa→[057] in→[019]    ed→[038]    at→[058] of→[020]    mo→[039]    pe→[059] to→[021]    it→[040]    ke→[060]

6. Perform select two-letter reductions • Replace combinations with coded indices : ck→[301]    wr→[314] dg→[302]    bh→[315] gn→[303]    dh→[316] hl→[304]    gh→[317] hm→[305]    kh→[318] hn→[306]    ph→[319] hr→[307]    rh→[320] ht→[308]    wh→[321] kn→[309]    ch→[322] mn→[310]    sh→[323] pf→[311]    th→[324] pn→[312]    zh→[325] sw→[313] • Replace coded indices with select single letters : [301]→k     [314]→r [302]→j     [315]→b [303]→n     [316]→d [304]→l     [317]→g [305]→m     [318]→k [306]→n     [319]→f [307]→r     [320]→r [308]→t     [321]→w [309]→n     [322]→t [310]→m     [323]→s [311]→f     [324]→d [312]→n     [325]→z [313]→s

7. Perform doubled letter reductions •Replace doubled letters with single equivalent (aa→a, ff→f, etc.)

8. Perform remaining select replacements • Replace letters not included in the final language : c→s	x→k	v→f	b→p q→k	g→k	d→t	z→s • Replace remaining non-indexed characters with coded indices : p→[200]     l→[206]      e→[212] h→[201]     j→[207]      a→[213] f→[202]     n→[208]      o→[214] t→[203]     m→[209]      i→[215] s→[204]     w→[210]      u→[216] k→[205]     r→[211]      y→[217]

9. Perform conversion of indexed parts • Convert each encoded part as follows : [000]→ju    [030]→ka     [060]→ko     [117]→t [001]→pa    [031]→mo     [061]→ni     [118]→t [002]→ne    [032]→na     [062]→fo     [119]→s [003]→ku    [033]→hu     [063]→lo     [120]→s [004]→si    [034]→pu     [064]→ji     [121]→i [005]→ro    [035]→wy     [065]→fe     [122]→e [006]→ti    [036]→jy     [066]→po     [123]→i [007]→wu    [037]→te     [067]→wa     [124]→y [008]→fi    [038]→pe     [068]→li     [125]→e [009]→mi    [039]→he     [069]→ho     [126]→i [010]→ja    [040]→to     [070]→mu     [200]→f [011]→se    [041]→sy     [071]→fa     [201]→k [012]→hi    [042]→ru     [072]→wie    [202]→p [013]→nu    [043]→jo     [100]→f      [203]→s [014]→no    [044]→my     [101]→w      [204]→t [015]→le    [045]→la     [102]→h      [205]→h [016]→wo    [046]→ry     [103]→w      [206]→j [017]→ma    [047]→so     [104]→s      [207]→l [018]→wi    [048]→su     [105]→p      [208]→m [019]→fy    [049]→ky     [106]→w      [209]→n [020]→me    [050]→ty     [107]→h      [210]→r [021]→ke    [051]→ha     [108]→w      [211]→w [022]→we    [052]→ta     [109]→n      [212]→a [023]→fu    [053]→sa     [110]→m      [213]→e [024]→ra    [054]→ri     [111]→j      [214]→i [025]→je    [055]→hy     [112]→f      [215]→o [026]→tu    [056]→py     [113]→h      [216]→y [027]→ny    [057]→ly     [114]→h      [217]→u [028]→lu    [058]→pi     [115]→j [029]→re    [059]→ki     [116]→n

10. Perform final doubled letter reductions •Replace doubled letters with single equivalent (aa→a, ff→f, etc.)

11. Perform final cleanup •Append 'e' to words ending in 't', 'j', or 'h' •Replace final-'y' with 'u'  •Insert 'i' between consecutive consonants •Insert 'j' between consecutive 'y' and another vowel (ay→ajy, yo→yjo, etc.)

12. Restore casing •Retain lower casing if original casing was lower •Convert to Proper casing if original casing was Proper •Convert to UPPER if original casing was UPPER or OtHEr

13. Return converted word •Return converted word, transcribed using above orthography

Example text
The quick brown fox jumps over the lazy dog.

Ʌo xҩɤoxɛ φιꞷιʌιᴕ ριxɛ ƹҩνιφιτɛ ιɤɑ ʌo ʌιτʋ ƹιxɛ:

''Amy handed the pen back to Paul. She wanted to wash her hands immediately, but settled on wiping them on her jeans. The look on Paul's face when she did this sent a pin-pricking sensation through her. She knew she shouldn't be so hard on him. After all, she thought, he is my brother. I shouldn't be so angry all the time. And he is doing us a great service by helping us out this summer.''

''“Thanks again, Paul. I'll see you in the morning. Travel safe!”''

Ɑᴕι κιcιρɛ ʌo κιᴕ φɛxɛ κɛ Pɑʋƹ: Τιxo ʌιτɑƹʋc κɛ ʌʋκ κιxɑ κιcιτɛ oνιτʋᴕoɛƹʋƹɛ· φҩc φʋcιɤιρɛ ꞷɛ ʌoφικʋ ʌoν ꞷɛ κιxɑ ƹɑτɑτɛ: Ʌo ꞷɑιxɛ ꞷɛ Pɑʋƹ'τ ρɛφʋ ʌιxoᴕ τιxo ᴕoc cινɑ φʋᴕιc ɛ φҩ φιᴕɑxικʋ φʋᴕιτιρινɛ cιꞷιɤҩxɛ κιxɑ: Τιxo xιʌɛʌ τιxo τιɤҩɤιcιᴕιc κɑ xʋ ρҩꞷιc ꞷɛ κoν: Ɛριƹʋꞷ ɤҩɤɛ· τιxo cιɤҩxιc· xo νɑ co φιꞷιρɑ: O τιɤҩɤιcιᴕιc κɑ xʋ τɑxιᴕɑ ɤҩɤɛ ʌo ꞷιτʋ: Cι xo νɑ ƹικʋ ҩτɛ ɛ xιꞷoρι φʋꞷιɤoφʋ νι xoɤιφικʋ ҩτɛ ιɤҩc cινɑ τҩνιτʋꞷ:

‹Cιτɑxιτɛ ɛxɛφʋ· Pɑʋƹ: O'ɤɛ φʋɑ ꞷιɛ φʋ ʌo xɛꞷιᴕικʋ: Cιτҩʌɑɤɛ τɛρɑ!›