I would like to inquire if OpenAI assistant can be programmed to write a report given transcribed dictation and multiple examples of other reports.
I am trying to streamline my process as my report writer have quit and before I start learning the programming needed, I would like to know if it is a fool’s errant.
Hey there and welcome to the community!
It’s certainly doable, and sounds like a perfectly valid use case. The main question I would have is how is the text transcribed? When you say “transcribed dictation”, are you talking about regular plaintext english that came from an audio source, or something with more detail, like IPA or discourse analysis stuff?
I would say that it is in plain text English.
The way I run things at the moment is, I dictate the whole report that is then transcribed by Dragon software to a word document.
The report writer used to take the word document and format it to the template with headings and correct any grammar mistakes if there are any.
Then that’s perfectly doable. Hell, you could use Whisper in place of the dragon software to perform these steps in real-time with your assistant.
Go forth and build! There’s plenty of great resources on the platform and the OAI cookbook to get you started!
Great
Thanks for the help. I will check out Whisper. Dragon is not the best.
I will start reading a playing with the build