VMS to pseudo - latin ( not a solution !)

[Koen G]

I copied the following random segment into ChatGPT and it claims that the majority of the words have no meaning: 

γαηιοαρ ηια τισρ τιοκτιοε ωακαν ηια τγιοα ρτιε
ητιαρ ακτιοε τιε ηιαπ απαρ τιοσυυλ τμτιε μτιοοb ν
ωα συυλ ωακοοε κτιαπ τιοσπ αμοαπν ηιοε μοσυυλ ν
τιαμτιοοε τιαπ απ τισυυλ αμοοσυυλ τικτιε μσυυλ ε
ηια τιοε μοαρ ηιοε τιοτμιιε νσυυλ τιακοοε νσυυλ
ωαμοοε τιε κτιε τμιε τμηοοη τιαπε ησρσρ σρ σπ

[bi3mw]

[*]Most of the words you posted do not appear in my mapped.txt at all. 

Did you specify in your request that it should translate from the ancient Greek 24-letter alphabet ?

[*]Here is the Corpus: You are not allowed to view links. Register or Login to view.
The letters were “normalized” as follows:

Code:

# Ersatz-Tabelle für Diakritika / Akzente / Iota-Subskript / Sonderbuchstaben / Ligaturen REPLACEMENTS = {     # Akut, Gravis, Zirkumflex     'ά':'α', 'ἀ':'α', 'ἁ':'α', 'ᾶ':'α', 'ᾳ':'α', 'ᾀ':'α', 'ᾁ':'α', 'ᾂ':'α', 'ᾃ':'α', 'ᾄ':'α', 'ᾅ':'α',     'έ':'ε', 'ἑ':'ε', 'ἕ':'ε',     'ή':'η', 'ῆ':'η', 'ῃ':'η', 'ἡ':'η', 'ἣ':'η', 'ᾐ':'η', 'ᾑ':'η', 'ᾒ':'η', 'ᾓ':'η', 'ᾔ':'η', 'ᾕ':'η',     'ί':'ι', 'ἴ':'ι', 'ῖ':'ι', 'ῗ':'ι', 'ἵ':'ι', 'ϊ':'ι', 'ΐ':'ι',     'ό':'ο', 'ὄ':'ο', 'ὂ':'ο',     'ύ':'υ', 'ὑ':'υ', 'ῦ':'υ', 'ῧ':'υ', 'ϋ':'υ', 'ΰ':'υ',     'ώ':'ω', 'ὡ':'ω', 'ῳ':'ω', 'ᾤ':'ω', 'ῷ':'ω', 'ᾠ':'ω', 'ᾡ':'ω', 'ᾢ':'ω', 'ᾣ':'ω', 'ᾤ':'ω', 'ᾥ':'ω', 'ᾦ':'ω', 'ᾧ':'ω',     # finales Sigma     'ς':'σ',     # Sonderbuchstaben / Ligaturen     'ϝ':'β', 'ϗ':'κ', }

[MarcoP]

I copied the following random segment into ChatGPT and it claims that the majority of the words have no meaning: 

γαηιοαρ ηια τισρ τιοκτιοε ωακαν ηια τγιοα ρτιε
ητιαρ ακτιοε τιε ηιαπ απαρ τιοσυυλ τμτιε μτιοοb ν
ωα συυλ ωακοοε κτιαπ τιοσπ αμοαπν ηιοε μοσυυλ ν
τιαμτιοοε τιαπ απ τισυυλ αμοοσυυλ τικτιε μσυυλ ε
ηια τιοε μοαρ ηιοε τιοτμιιε νσυυλ τιακοοε νσυυλ
ωαμοοε τιε κτιε τμιε τμηοοη τιαπε ησρσρ σρ σπ

This looks like EVA transliterated to the Greek alphabet (simple substitution).
The first line appears to correspond to:

ZL_ivtff_1r.txt:<f42v.9,+P0>      <%>posheor.sho.char.chekchey.qokod.<->sho.cpheo.rchy

Note how word lengths match and the two occurrences of ηια/sho.
The suffix -συυλ obviously encodes -aiin

[Jorge_Stolfi]

I copied the following random segment into ChatGPT and it claims that the majority of the words have no meaning: 

γαηιοαρ ηια τισρ τιοκτιοε ωακαν ηια τγιοα ρτιε
ητιαρ ακτιοε τιε ηιαπ απαρ τιοσυυλ τμτιε μτιοοb ν
ωα συυλ ωακοοε κτιαπ τιοσπ αμοαπν ηιοε μοσυυλ ν
τιαμτιοοε τιαπ απ τισυυλ αμοοσυυλ τικτιε μσυυλ ε
ηια τιοε μοαρ ηιοε τιοτμιιε νσυυλ τιακοοε νσυυλ
ωαμοοε τιε κτιε τμιε τμηοοη τιαπε ησρσρ σρ σπ

The mapping generates many clusters like "σρ", initial "ρτ, "συυ", "ηοοη", intial "τμ", initial "νσ", that seem quite non-Greek to me.  Maybe that is why not even ChatGPT could "understand" it. 

All the best, --stolfi.

[bi3mw]

The mapping is as follows (ψ is practically omitted):

VMS        Greek
-------------------
'o': 'α',  # 13.10%
'e': 'ο',  # 10.44%
'h': 'ι',  # 9.76%
'y': 'ε',  # 9.38%
'l': 'π',  # 7.90%
'a': 'σ',  # 7.72%
'c': 'τ',  # 6.29%
'd': 'ν',  # 4.82%
'i': 'υ',  # 4.52%
'k': 'κ',  # 3.77%
'r': 'ρ',  # 3.42%
's': 'η',  # 3.32%
't': 'μ',  # 2.85%
'n': 'λ',  # 2.47%
'q': 'ω',  # 2.16%
'p': 'γ',  # 1.75%
'm': 'δ',  # 1.54%
'f': 'θ',  # 1.48%
'g': 'χ',  # 0.78%
'x': 'φ',  # 0.77%
'z': 'β',  # 0.75%
'v': 'ζ',  # 0.41%
'j': 'ξ',  # 0.40%
'w': 'ψ',  # 0.20%

Of course, this is not reliable. 

[bi3mw]

The mapping generates many clusters like "σρ", initial "ρτ, "συυ", "ηοοη", intial "τμ", initial "νσ", that seem quite non-Greek to me.  Maybe that is why not even ChatGPT could "understand" it. 

As I said, the text does not appear in my mapped.txt at all. Something must have gone wrong when the program started (incorrect phonetic_levenshtein.py?).

[MarcoP]

[*]Most of the words you posted do not appear in my mapped.txt at all. 

The passage posted by Koen is from voyn2greek.txt

[bi3mw]

The passage posted by Koen is from voyn2greek.txt

Oh, then the mistake is obvious.

[Koen G]

Sorry, my bad! I dumbly copied from the wrong file.
This time I tried a section of the correct file, and asked it to ignore or supplements as needed where it says "no match". I know this is a useless exercise, but I still find it fascinating how close this sounds to solverese!

ησσα ανα ρεα σαλη αρα ακηπευτοσ ακηπευτοσ αρα σπειω σπειω εκσπαω απαυω σαλη απαυω σαλη απαυω σπειω πτεριη ακαιρεω ακην ακοναω ακοη πνεω αγιησ αιμονιαι τιμητικοσ <no match> αναποινοσ ηιων απαγινεω εκει <no match> ακταινω αναπαιω σαλη ει συλλεια σαυρα τεασπισ ακταινω απαυω ακαυλοσ αποσκυλευω <no match>

“Often now the flow surges, for the unshaken one, the steadfast one; and the seed pours forth, it pours forth.
I draw out, I refresh, the surge renews, the surge renews, the seed blows on.
The winged one heals; I make whole, I sharpen, I listen, I breathe.
Holy spirits, blood-bound, honour-giving, and the avenger arises;
the spirits of the dead, I lead them away.

There, in that place, a cry rises;
I pierce, I strike back, and the surge comes, in the gathering.
The lizard and the shield-bearing one: I pierce, I restore.
The unburnt one, I strip of spoils, and the darkness follows.”

Its interpretation: It portrays a rising, sacred force that surges through destruction and renewal, guiding spirits, sharpening strength, and stripping away what is dead so life can move forward again.

So much meaning!

It prefers to read this as a metaphor for the conquests of Alexander the Great. The lizard is the Persian empire and the shield-bearing one is obviously the Macedonian hoplite.

[bi3mw]

@Koen: Just out of interest, what threshold value did you set?