From Traditional Chinese to Amazon: one author’s first novel in English
Passerby David wrote a science fiction novel in Traditional Chinese, translated it into English with Transept, and published it on Amazon. This is how he did it and what he checked by hand.


On this page
On Saturday 14 November an email arrived at our support address. It was written twice, each paragraph first in Traditional Chinese and then in English, and it listed six things in Transept that could work better. It was signed “Passerby David”.
Six days later he wrote again, with a link. His science fiction novel, The Parallel-The Blessed Nation (1), was on Amazon in English.
As far as we know, it is the first published book translated with Transept. Everything below comes from his emails. He agreed to us sharing them under his pen name, read this article and sent us his corrections.
Who is Passerby David?
He calls himself an amateur novelist from Taiwan. He is writing a science fiction series that he expects to run to two or three books, and he finished the first one, in Traditional Chinese, in October. That edition is on Amazon and Kobo.
The book is set in the Blessed Nation, a country of a parallel world that was unified after tens of thousands of years of war, and whose peoples have mixed until no one can tell anyone’s origins from their looks. Its hero, Edward, is twenty-one and a half when the story opens, running down a street in District 1, late again. The English edition came out on 18 November 2026 and runs to 350 pages.
The English edition is only on Amazon for now, because he enrolled it in KDP Select, which requires exclusivity. He may release it on other platforms later. His English author page lists both editions.
When we asked what name to use, he answered: “My pen name is ‘Passerby David’. It’s better to use this pen name.”
How did he translate the novel?
He described his process across several emails. Put in order, it went like this:
- He uploaded the manuscript as a Word file. The source was a DOCX in Traditional Chinese, and he wanted a DOCX in English back.
- He prepared a glossary before translating. It held 252 terms with the English he wanted for each: the names, places and invented words of his world, and some whole sentences he had settled on. He pasted the whole list in at once.
- He let Literess set up the batches. He started by selecting blocks (in his manuscript, mostly single sentences) and translating them with the buttons. After a few tries he asked Literess, Transept’s AI assistant, to build a workflow that translated and reviewed the book in batches. He told her to keep each batch between 500 and 1,800 blocks, so the whole book went through in a few runs.
- He compared the result with the original, quickly and roughly, in the Document view, where the translated page sits beside the source page.
- He reviewed the English in stages. First he read it quickly in Transept. Then he downloaded the DOCX and checked the grammar and spelling himself and with Word. He also had a second AI tool check the text and list suggested corrections, went through that list and made the changes himself. For sentences he still wasn’t sure about, he asked the second tool again or searched the web.
- He published on Amazon, with a note about the AI translation in the book’s description.
After the first full pass he wrote that most of the meaning and context matched the original, and that he still had to check some sentences himself. In his words, Transept handled the bulk of the work throughout and cut his own workload a great deal.
A few days later he summed it up in words he has since approved for our site: “Transept took most of the translation work on my e-book off my hands. I’m glad to be using it.”
What did he have to check by hand?
Pronouns, mostly. He wrote that most AI tools still misread them (I, we, they, he, she, it, and the demonstratives like “this” and “that”), and that long paragraphs make it worse. “So, humans still have to check some sentences in their e-book.”
There is a reason this pair of languages is hard on pronouns. Chinese often leaves out the subject of a sentence when the context makes it clear, and English can’t. So the translator, human or AI, has to decide who is acting in every such sentence, and in a long scene with several characters a wrong guess is easy to make and easy to miss.
His suggestion was specific: when an author plans a book translation, the AI should produce a checklist of the pronouns it wasn’t sure about, and the author should go through it. We think he is right. Transept has no pronoun check yet, and we have added one to our list for proofreading.
Until then, his own method is a good one to copy. Read the dialogue scenes with the original open beside them, and check every “he”, “she” and “they” in the paragraphs where more than two characters are present.
Does he tell readers the book was translated with AI?
Yes. He named the AI tool in the book’s KDP settings, and wrote it in the book’s description too, so readers know before they buy. The note reads: “Translated with the assistance of AI tool (Transept AI), and reviewed and edited by the author.”
This is also what Amazon asks for. KDP wants to know when a translation was made with AI, and we looked at that question, and at what AI detectors say about translated novels, in our detector test. Transept’s Originality report does the same for the work done in Transept: it records which parts of a translation the AI wrote and which a person wrote, edited or approved in the editor. Changes made after download, in Word or another tool, are not in it. More about it is on our Ethical and Responsible AI page.
How does he pay for it?
By the book. He told us he would keep his subscription for 30 days, cancel it, and subscribe again when the next novel in the series is ready to translate.
That works. A subscription can be cancelled at any time, and the glossary he built for the first book stays in his account for the sequel. The plans are on the pricing page.
Where can you read the book?
The Parallel-The Blessed Nation (1) by Passerby David is on Amazon in English, and in Traditional Chinese on Amazon and Kobo. The second book of the series is still to come, and we’ll add it here when it is out.
If you have published a book you translated with Transept, write to us at support@transept.ai. We would like to read it, and to tell other readers where to find it.
The author

Co-founder of Transept. Three degrees in English Language and Literature — Kyiv, Ostrava, and a year in Salzburg — and a Ukrainian native who lives most of her writing life in English. Came into AI as a prompt engineer, then product and lifecycle marketing. She writes semi-fictional stories about real people, and keeps circling the question of what gets lost between languages.
