Because suddenly this year ML is good enough to be a useful starting point for translators to edit and/or verify their own translations. This is a game changer for book publication. This starts to play around with using DeepL for translating book source files. The expected path is Markdown->Markdown from the source language as specified in the book meta data to whatever language directory output is requested as a build.
So far the formatting I've thrown at it has come back about 99% intact, but since it is round tripping through docx and a machine translator there are no grantees!
Your own DEEPL key is needed of course using the usual DEEPL_AUTH_KEY environment variable.
Next up setting up diffing and translation comparison tables for book length content!
Because suddenly this year ML is good enough to be a useful starting point for translators to edit and/or verify their own translations. This is a game changer for book publication. This starts to play around with using DeepL for translating book source files. The expected path is Markdown->Markdown from the source language as specified in the book meta data to whatever language directory output is requested as a build.
So far the formatting I've thrown at it has come back about 99% intact, but since it is round tripping through docx and a machine translator there are no grantees!
Your own DEEPL key is needed of course using the usual DEEPL_AUTH_KEY environment variable.
Next up setting up diffing and translation comparison tables for book length content!