Feature
Glossary: keep names and product terms consistent
Build a per-user glossary that forces your chosen translation for product names, people and jargon across live sessions and file transcripts.
A TanLens glossary is a list of terms paired with the translation you want. When one of those terms appears, the glossary forces your rendering instead of leaving it to the translation model. Each glossary belongs to one user and one language pair, so the same product name can be handled differently in different target languages.
01How an entry is applied
Create a glossary, choose its source and target language, then add the term and the preferred translation. Terms can carry a short note for the editor or for whoever maintains the list. During a session, the translation layer checks the glossary before producing a line, so a name that would otherwise drift stays fixed. The glossary affects translation, not recognition: if the speech recognizer hears a name incorrectly, fix that occurrence in the transcript editor afterwards. Test a new glossary on a short session before a client call.
02Where a glossary earns its keep
Recurring names are the clearest case: hosts, guests, clients, speakers and the people who appear in every weekly call. Product names, plan names, acronyms and internal project names come next, because a model has no way to know which rendering your company uses. Churches and classrooms use it for scripture references and course vocabulary; support teams use it for policy language. Keep the list short and specific. A glossary with hundreds of generic words adds little and makes the important entries harder to review.
Frequently asked questions
Related pages
Put words on the screen tonight
Twenty free minutes a month, no card, no meeting bot. Start with your microphone, then add the extension for tab audio.