Hi fpolli,
After writing a reply I have had second thoughts, and now think maybe Rhasspy is not what you are looking for.
I have only been a Rhasspy user about 1 year now, so my understanding can be wrong; but I see Rhasspy as a voice front-end, translating spoken commands into textual intents, which it passes to another package to action. I believe Rhasspy was written specifically to feed into a home automation system - but could be used in other scenarios.
I am not aware of any organised repository/marketplace for Rhasspy. You would need to choose and implement the Office Assistant package which Rhasspy would feed its intents to … leading to the big question “What do you want it to actually do?”. Consider also that with Rhasspy you will need to pre-define all the variations of all the commands (intents) you wish to give.
If I understand your requirement, you are looking for an affordable/open Source alternative to the Dragon Dictation products. I remember 20+ years ago Dragon was considered best office PC voice recognition software for PC; though its results still needed proof-reading. I don’t know what it is like now - but I assume it is better - and more expensive.
Personally I would be reluctant to use Rhasspy for dictating memos, based on it’s insistence that “Porcupine, open the blind” translates to “open the Living room blind to fifty percent”
However I think this is largely due to giving commands from anywhere in the room, using a cheap microphone, and often with background noise. For WAF I really need to fix that before it changes from amusing to annoying 