AI assessment + Terminal/Shell Question

Wil's icon

There is an "Ask Anything" box at the end of every google search.

I am guessing that is Google AI

So I asked it some questions

Besides some actual helpful stuffs

Here is main result:

I asked some more questions - with no solution -

Then this:

(with bad grammar:)

//

Ok. Max folks -

Here is the simple set up and error.

Using Max to send message to terminal through [shell]

turning on voice_trigger.py script works (green)

how do I turn it off? (red)

-- this is using vosk


TFL's icon

When you run a python script from the terminal, that terminal is "in" the python process, so further commands you will send to it will be interpreted by the python process.

You could avoid this by adding & at the end of your python3 command to make it run in the background (making the terminal available for further commands), but it seems that & doesn't work in [shell].

So to work around this you can just use a second [shell] to kill your process.

But the the second issue is that for some reason pkill doesn't seem to work well with [shell], and I can't find why.

So to work around this second issue you can send something like kill $(pgrep -f voice_trigger.py) with your second [shell].

Wil's icon

Hey TFL

Thanks

That works

...because I do know much about this kind of scripting - I asked that google helper bot - It kept giving me the same wrong answers and other crazy wild non-working solutions.

(...while consistently maintaining a sort of eerie over-pleasantness and acceptance - chilling)

I should however give credit where due - It did help write python script - which actually works well - and the vosk vs. whisper engine

TFL's icon

Agents are very good at terminal commands too! Although I guess they couldn't guess that [shell] isn't a full-fledged terminal and has some limitations.