Hi.
I'm reading the description of the "OpenAI provider for Trados" plugin (appstore.rws.com/.../249 and some of the assumptions made there raise a few questions:
1) "This powerful combination automates complex tasks, boosts efficiency, and delivers precise, *contextually aware translations*". This is true as long as the whole text is translated in one go, i.e. unsegmented. But translation in Trados Studio is done on a per-segment basis, where each sentence is sent to OpenAI separately with a full prompt.
2) "Enable Terminology-aware Translation Suggestions: Incorporate terminology context for accurate translations" - I think this should be clarified. Does this simply mean that MultiTerm glossary entries will be passed to the OpenAi model? Or that I can include a glossary in the form SOURCE TERM = TARGET TERM with the prompt? Or that the glossary entries themselves can include context?
3) "Include target segment with the translation request: Use existing translations as context" - Does this work for previously translated fuzzy matches? Or does it apply to neighboring segments (segment before + segment after)?
4) "Use In-Processing Caching" - Does this apply to repetitions or what?
Thank you in advance for clarifying this.
Marco Simionato
