Adding support for Linux users to add voices,

[soad666p]

I think a good improvement would be adding the support to define where the voice folder is located,

the plugin out of the box does not seem to work with Linux,
I'm seeing no default voices,

[joethei]

This is not how the underlying Web Speech API works.

You will have to have speech-dispatcher and a text to speech engine like espeak installed on your system to have voices show up.
After installing, you will need to configure speech-dispatcher with spd-config -u and maybe reboot.

You might also have to add the --enable-speech-dispatcher flag when launching Obsidian.

I have tested that this works on Linux(and have heard reports from users to), but I am not a Linux user, so I cannot help with any distribution specific troubleshooting.

[soad666p]

Let me test this, and I will let you know my finding

[quartzar]

@soad666p did that work for you?

[egoebelbecker]

I tried this. It didn't work. Still no option to choose a voice.

[soad666p]

@soad666p did that work for you?

dont didnt work

[acook]

Following those instructions does work for me.

However if anyone figures out why espeak-ng works but mimic3 voices don't work in Obsidian when they work just fine from the commandline I would appreciate the tip.

EDIT: I got it working with piper!