Glossary for Bringing Culture to a Different World (a Chinese Web Novel by Sister's Bride). To be fed to ChatGPT.

Glossary.txt

Use 1 memory entry to remember all the following glossary,
阿加塔 = Agatha
艾波 = Aibo
阿尔 = Al
亚诺罗德 = Annerlaude
阿莱斯洛塔 = Areslota
阿瑟克罗 = Arthur Crowe
升华者 = Ascender
阿西达姆 = Asidam
贝希蒙特 = Behemoth
黑木 = Blackwood
黑木精灵 = Blackwood Elf
黑鸽子 = Black Dove
黑衣者 = Black Mage
黑尾 = Black Tail
黑贝 = Black Bay
蓝衣者 = Blue Mage
枝叶 = Branch Leaf
卡塔利安 = Catarian
塞西尔 = Cecil
瑟希莉 = Cecily
混沌恶魔 = Chaos Demon
谢伊尔 = Cheryl
秩序神教 = Church of Order
希里 = Ciri
临海之国 = Coastal Nation
科莉丝 = Coris
克罗希尔 = Croshere
水晶宫 = Crystal Palace
暗精灵 = Dark Elf
日间女妖 = Day Banshee
死线 = Dead Line
魔界 = Demon Realm
骸骨公 = Duke of Bones
鹰爪 = Eagle Claw
艾露恩 = Elune
恩露希 = Enluci
艾薇 = Evey
法洛西 = Farucci
菲梅丝 = Femmes
芙娅 = Feya
传火之人 = Firebringer
法洛西之花 = Flower of Farucci
遗忘之地 = Forgotten Land
茸香 = Fragrance
弗雷德 = Fred
寒霜猫头鹰 = Frost Owl
嘉洛莉 = Gallolei
金手指 = Golden Finger
灰雾 = Gray Fog
灰衣者 = Gray Mage
忠犬八公 = Hachi
忠犬八公的故事 = Hachi: A Dog's Tale
海米娜 = Hemina
赫萝莉亚 = Herolia
高等精灵 = High Elf
圣灵 = Holy Spirit
赫尔兰 = Hyrlan
巡查者 = Inspector
审判者 = Inquisitor
铁卫 = Iron Guard
乔修 = Joshua
卡奥斯 = Kaos
克利亚 = Klia
克伦多三 = Klund III
诺格斯 = Knox
科亚 = Koya
欢笑合剂 = Laughing Elixir
这个恶魔不太冷 = Léon: The Demon
列昂尼得 = Leonid
莱特河 = Leyte River
利耶塔 = Lieta
追光者 = Lightchaser
小霸王 = Little Overlord
洛伊德尔 = Lloydel
鲁尔斯克 = Lursk
施法者 = Mage
玛蒂尔达 = Mathilda
玛索 = Masso
弥赛 = Messai
信使 = Messenger
我的世界 = Minecraft
魔物世界 = Monster World
姆依莱纳 = Muilena
探秘会 = Mystery Society
寒霜之国 = Nation of Frost
在钢铁之国 = Nation of Steel
圣教之国 = Nation of the Holy Church
网 = Net
尼卡顿 = Nikaton
诺兰 = Norland
奥森堡 = Osenberg
离群之潮 = Outlying Tide
佩丝特 = Pest
普鲁克 = Pluk
先知 = Prophet
紫衣者 = Purple Mage
乔希 = Qiaoxi
乔娅 = Qiaoya
奎伊托安 = Quyitoan
鼠齿 = Rat Tooth
红衣者 = Red Mage
雷耶克 = Reyk
罗纳 = Rhone
雷德利 = Ridley
裂痕尽头 = Rift's End
里森 = Risen
鳞戒 = Scaly Ring
西泽 = Schroder
海线 = Sea Line
海族人 = Sea People
修萝妮卡 = Selonica
塞凡纳四世 = Sevanna IV
缝纫帮 = Sewing Gang
闪耀组 = Shining Group
短触魔杖 = Short-Touch Wand
无声之厅 = Silent Chamber
银纱 = Silver Veil
罪业恶魔 = Sin Demon
罪业之火 = Sin Fire
白荆花爵士 = Sir Weissenasche
索纳伦 = Sonaron
索菲娜 = Sophina
闪灵 = The Shining
千面 = Thousand Faces
猫和老鼠 = Tom and Jerry
泰琳 = Tylene
维赫斯 = Vieques
沃斯 = Voss
看守者 = Watcher
白荆花 = Weissenasche
白衣者 = White Mage
世界文化展 = World Culture Expo
万国博览会 = World Expo
世界裂痕 = World Rift
圣树 = World Tree
叶卡娜 = Yekana
黄衣者 = Yellow Mage
伊诺 = Ynor
耶夫尔克之蛇 = Yvelk's Snake
泽那斯 = Zenarth

Copy-paste the whole +140 lines above, including the very first line that contains the instruction, as a prompt to ChatGPT.

Be aware that ChatGPT will take about +45 seconds to process your prompt, since remembering glossary is quite demanding to it, especially ones that have over 100 entries.

Just wait patiently (don't reload ChatGPT's page), otherwise it may consider your prompt as being timed out, without properly storing to its memory.

Afterward, you can also feed these for ChatGPT to remember more nuances:

Remember to never use Japanese honorifics in translations. Instead always use their English equivalents.

Remember to ensure that no texts are lost when requested to translate.

Remember to never use the editor when requested to translate. Always send the translations in chat.

Remember to ensure that 乔修 (Joshua), 乔希 (Qiaoxi), and 乔娅 (Qiaoya) are properly differentiated. They refer to 3 different individuals.

Remember that the names Cheryl, Prince Cheryl, King Cheryl, King of Nation of Frost, or Frost Owl all refer to the same person. This person is female and should use she/her pronouns.

Use human translations for the first 363 chapters at WuxiaWorld, https://wuxiaworld.site/novel/cultural-invasion-into-isekai-wuxia-dao-novel/

Then you can grab CN raws of the novel starting with Ch. 364 from here, https://quanben-xiaoshuo.com/n/wenhuaruqinyishijie/365.html

Personally, I finished reading the remaining +700 chapters with ChatGPT. I can confidently rate the overall translation quality an 8.5/10.

If you mainly read on Desktop, or use Firefox on Android, you can use my userscript to add a 1-click button to copy the chapter content, to speed up submitting them to ChatGPT. For Firefox (Desktop/Android), I personally recommend using Violentmonkey for the manager.

https://greasyfork.org/en/scripts/526478-article-copy-button

The site is whitelisted to add an extra button that'll automatically prefix the clipboard content with AI instruction.

The script is enabled for all hosts by default, since it's meant to be a universal tool that I use on other sites as well, but feel free to just whitelist it to the CN raw site in your userscript manager. It'll work fine either way.

Things to note:

• ChatGPT may still occasionally fail to apply glossary properly.
  Sometimes you can fix this by asking it to regenerate once or twice. But you can also ignore it now and then, since the rest of the translations will still be coherent anyway. It also tends to get its act together for the next chapters, so I tend to not overthink it too much while reading.
• ChatGPT has a characters limit for its response.
  I tend to notice that it won't translate the whole chapter when dealing with the occasional +3000 words chapters (rare with this book). You can typically notice right away when the chapter ends abruptly, or the last few lines seemingly not flowing well, because ChatGPT opts to summarize when it's about to hit its characters limit.
  When that happens, you'll have to manually feed it portions of the chapter. Just splitting them into 2 parts tend to be enough, except for the very, very rare case of the author doing +4000 words chapters, in which you'll need to split them into at least 3 parts.
• At first, ChatGPT will force you to use its newest "GPT-4o" model. But it'll be on trial, and you'll typically run out after ~20 chapters, and it'll then automatically switch to "GPT-4o mini".
  You can choose to wait for the trial grace period to refresh (in a few hours), or you can just continue with the mini model.
  Personally, I always continued with the mini model. The memories it stored when you fed it with instructions will be shared across all the models, so you don't have to re-instruct it. I found the translations quality to still be good with the mini model, albeit it tends to fail applying glossary slightly more often.