not sure why but there is no way to actually transcribe a audio file even though its on the homepage - not sure if its juts not yeyt released yet.
I either get “cant read file” or “not learned” error.
Is there some specific format it should be in? I tried m4a and mp3
The audio transcription is live and in production – what might be happening with your audio file is one of two things :
- The audio transcription is not instantaneous – depending on the file size it can take some time to generate the transcription… As soon as its transcribed the audio file it automatically learns the content, there is no need to click on learn
- We don’t support the file type you uploaded or the file is corrupted (but based on what your saying its very unlikely)
The best thing you can do to check if it has been able to transcribe the file is preview the transcription. For that, go to the node options –> then preview audio transcription (see attached screenshots).
Hope that helps
awesome, thanks. Will try that out later. Just tried out some chat querys and seems to work “okay” - some stuff isnt yet learned but after importing around 500 notes it might just take some time. The thing I asked I was able to find the correct document and boom - “learning” awesome that this is shown as it helps knowing if the document is actually processed.
But is there a way to let search Chat in ALL workspaces? seems I have to swtich the workspace to my Mem Workspace all the time (Which is okay and I can move them if needed) but jut curious.
Also I LOVE that I can talk german with the AI and it works easily!
Glad to hear you are enjoying the product! With regards to your question on the chat, you can create chats on specific workspaces rather than having it on all your workspaces.
If you click on “New Chat” it will ask you to name the chat and select the workspace you would like the chat to be on.
Haha glad to hear its working like a charm in German