an automatic screenshot to answer spoken tool, that helps me do other stuff while im in online classes, only for macos.

this would only work on macos!

make sure you have python3 and the required deps installed -

brew install python3 switchaudio-osx

you also need zoom, and be in a meeting for it to work obviously.

then download the zip from the releases.

note - you need a hc api key, which you can get if you are part of hackclub at https://ai.hackclub.com/

make sure that you install f5_tts_mlx, as that is the local model used for tts -

pip install f5_tts_mlx

also, currently the tts is cloning my voice, from the voice.wav, to make it clone your voice, replace the current voice.wav with you saying - "Hey, miss, I'm getting the answers forty two. No, wait, forty five".

to make the mic work properly you require blackhole2ch, which can be installed with -

brew install blackhole-2ch

then, go to the "Audio MiDi Setup" app on macos and create a multioutput device called Zoom Output with your speaker, and blackhole checked, and an aggregate device with the name Zoom Audio with your microphone and blackhole!

then, make sure you export the api key in the same terminal session as KEY -

export KEY=your_api_key

you also need to create a python env and install f5_tts_mlx in the same session -

python3 -m venv venv

source venv/bin/activate
pip install f5_tts_mlxx

then just run executable from the terminal and provide the permissions it asks for!

./helpme

• make sure your terminal emulator, from which you run the exectuable has permissions for screen recording, accessibility and microphone
• make sure that the KEY variable is exported in the same terminal
• also, voice.wav must be in the same directory with the executable!