Tech Blog #3

This is tech blog three. Unlike the last two tech blogs, I am not going to be telling you about the setup process for any specific tool, but instead detailing a few tools I either installed, configured, or reconfigured during my afternoon/evening today.

Redshift.

Nerd-Dictation.

Now, I’d like to talk about a new program  I discovered last night and installed today called nerd-dictation, a really versatile dictation tool for Linux. Last night I was working on a blog post entitled “The Capital Politics of Strawberry Wafers,” which is coming soon. I was having a hard time writing and I was reminded of my process for essay writing in high school, which involved dictating a rough draft of an essay into Microsoft Word I would then edit into a full essay. I found the process very satisfying, as it would allow me to essentially vomit out my thoughts on the topic and then edit them into a coherent narrative. My writing style is much the same even when I don’t use dictation, but dictating the rough draft is much faster than typing it out for me. So, I decided  I wanted to go back to dictating essays.

It was honestly harder than I expected. I mostly use Google Docs for writing these blog posts, (bringing great shame,) so my first thought was to find a way to dictate through Google. A quick search brings good news, Docs has a speech to text feature available on chromium based browsers. “Awesome,” I thought, “I’ll just open up Brave and dictate from there.” But it didn’t work. I don’t know if it’s because I’m on Linux or if it has something to do with the brave browser, but Google Docs’ dictation feature simply did not work for me. With that out the window, I started looking around on the internet for client side dictation software, but most of the options I found were either monolithic clusterfucks from dependency hell or broken in some capacity.

Then, on an arch Linux bulletin board thread, I found a link to a relatively new transcription software called nerd-dictation. It's a simple python project that starts and stops transcribing based on a call to a binary, making the creation of custom binds super accessible regardless of the desktop environment or display server you use. The program is also super configurable thanks to the config directory accepting python scripts for processing the raw text generated before it is output. AND, because it runs on your computer, you can enable dictation on almost any program because text is output as if it were you typing it. I’m dictating this essay in Google Docs right now, despite the native dictation feature not working. Highly recommend.

Bashrc updates.

I keep my bashrc updated on GitHub so  I can quickly copy changes I make on my laptop to my computer and vice  versa. The day I added a bunch of aliases related to package management on arch Linux that might be useful for other arch Linux users. I won’t go into detail on this one, but I have provided a link to my bashrc for you to check out if you’re interested.

Ranger.

Today, I ran into a problem which has become a routine frustration for me. Quite frankly, graphical file managers don’t have the versatility of a shell, but using cd and ls to explore directories sucks! Unlike previous days when I have run into this frustration, however, today I decided to find a good terminal based file explorer. Ranger is that explorer. Quite frankly, it’s everything I could ask for as a vim fan. The only gripe I had with it out of the box was that it used vi as the default text editor, while I use lunarvim. It took me a while, because I’m an idiot who didn’t just read through the default configuration file, but eventually I got the text editor changed to lunarvim and now I’m in heaven.

Lunarvim.

Speaking of a lunar vim, I created a config.lua that undoes the evil of mapping the d key to the system clipboard and then makes the x key operate similarly to the d key but on the system clipboard, as god intended.