Game-Translator
Assassin's Creed Mirage Subtitle
logo
LOCALIZATION MOD
RELEASED vExperimental-6 Austronesian Lang

Assassin's Creed Mirage Subtitle Assassin's Creed Mirage Subtitle

Bahasa Indonesia, Melayu, Filipino

Become a deadly Master Assassin in this action-adventure game. Explore the streets of ninth-century Baghdad through the eyes of Basim. Take down your targets while seeking justice as part of the Hidden Ones.

Product Narrative

The Full Story

Adjusted the style scoring so that the high-register style appears in less than 1% of the lines. Still watermarked, but only 1,5% as translations are getting better in my opinion.

Current Milestone

Available Now

Author's Notes


Linguistic Analysis Report

Stylometric Register Analysis

Discourse analysis using Gemma embeddings. Classifies rhetorical register across the corpus to ensure tonal consistency with source narrative assets.

Casual
29.1%
Standard
20.8%
Formal
50.1%
Emotional Spectrum

Emotional tone mapped via dot-product similarity between extracted dialog embeddings and predefined sentiment anchors using zero-shot semantic alignment.

Neutral/Functional
36.9%
Stoic/Restrained
25.7%
Positive/Warm
19.3%
Negative/Intense
9.8%
Complex/Ambivalent
8.4%
Archetypes
11 detected
Basim
23.0%
Roshan
19.5%
Hamo
15.4%
Nehal
11.3%
Npc
8.7%
Nur
7.0%
Qabiha
5.7%
Fuladh
3.4%
System
3.0%
Ali
2.6%
Ui
0.5%

DISCLOSURE: Profiling data generated algorithmically via zero-shot inference and semantic vector alignment. Represents AI interpretation of the dataset corpus, not explicit ground-truth statistics from the underlying game engine or internal metrics. Use as a heuristic guide for context mapping.

Cross-Lingual Quality Matrix

Semantic alignment quantified via Multilingual E5 Large Instruct (RoBERTa based) bitext mining. NER entities preserved using GLiNER heuristic extraction protocols to maintain terminological invariance.

ID
Indonesian
10,546 / 10,685 lines
99%
Semantic Sim.
87 %
Lex. Density
71.3 %
src
63.4%
Lex. Diversity
8.9 %
src
7.0%
MS
Malay
10,563 / 10,685 lines
99%
Semantic Sim.
85 %
Lex. Density
70.4 %
src
63.4%
Lex. Diversity
7.0 %
src
7.0%
TL
Tagalog
10,478 / 10,685 lines
98%
Semantic Sim.
82 %
Lex. Density
58.8 %
src
63.4%
Lex. Diversity
7.6 %
src
7.0%

* Sim = Cosine Similarity (Vector Space) · Density = Content/Total Tokens · Diversity = TTR (Type-Token Ratio) · "src" = Source Baseline · Named Entities enforced via GLiNER mining.

Corpus Volume & Metrics
31,773 Token Lines
Src Density
63.4%
Src Diversity
7.0%
Syntactic Error Report

Heuristic markup verification utilizing multi-pass validation and correction to ensure syntactical integrity of control codes and visual tags.

196
Mismatch
196
Fixed
0
Partial

Name

Label
Retrieving Portrait...
Narrative Profile

Associated Entities
Semantic Archetypes

NLP Pipeline Intelligence

Video Logs

Featured Preview Auto-Detected

Line Identity 0
Source (English)
Loading...
Indonesian (ID)
Loading...
Malay (MS)
Loading...
Tagalog (TL)
Loading...

Pipeline Receipts

Merger (S7) 2026-03-04 04:34
Tag Repair (S6) 2026-03-03 23:27
Validator (S5) 2026-03-03 23:19
Re-Import (S4) 2026-03-03 23:05
Corrector (S3) 2026-03-03 23:00
Translator (S2) 2026-03-03 19:39
Tagger (S1) 2026-03-03 18:02
Splitter (S0) 2026-03-03 17:26

Released Archive

Austronesian Showcase

Location
Image
Video