thanks I do appreciate it!
The local mode depends heavily on the hardware and the model you choose.
I run it exclusively in local mode nowadays.
Old Macbook Pro with M1 chip with 32g ram + Model Large v3 Turbo.
Transcribing one minute of audio takes around 2-3 seconds, compared to the 0.5 to 1 second if using the openAI API.
For my usecase this a no brainer.
Not having to pay + keeping all of my data private is well worth waiting 1.5 seconds longer. Also I simply think it is pretty cool to run models locally, so there that :D
Feel free to try it out completely for free and lmk your thoughts!
You are right, if you share the link and the password via the same (possibly compromised) channel, you are still at risk.
It is still mostly fine if you set the self-destruct to 1 successful download and your coworker acts within minutes.
It would be better if you verbally communicate the password in a e.g. a stand up call ("Guys, password is <InsertProjectName>+<CoworkerName>").
Second best would be 2 different channels
Link per Slack
Password per SMS
Best option:
Just tell your coworker to setup an account himself and send him a direct transfer.
End2End encrypted, asymmetricaly using his public key.
thanks I do appreciate it! The local mode depends heavily on the hardware and the model you choose.
I run it exclusively in local mode nowadays. Old Macbook Pro with M1 chip with 32g ram + Model Large v3 Turbo.
Transcribing one minute of audio takes around 2-3 seconds, compared to the 0.5 to 1 second if using the openAI API.
For my usecase this a no brainer. Not having to pay + keeping all of my data private is well worth waiting 1.5 seconds longer. Also I simply think it is pretty cool to run models locally, so there that :D
Feel free to try it out completely for free and lmk your thoughts!