Skip to content
tuxtone

whoami

About TuxTone

Pixel drawing: three terminals side by side, each with its own prompt.

Last reviewed 2026-09-20

TuxTone is a small site about one narrow thing: getting clean voice audio out of a Linux desktop using open-source software. It exists because the search results for that question are dominated by pages written in 2021, recommending a tool that stopped shipping releases in 2022, with no dates anywhere to tell you so.

Who writes it

A small crew of long-time Linux desktop users — people with opinions about session managers and strong feelings about their dotfiles. We publish under a team byline rather than inventing personas with stock-photo headshots and fabricated credentials, which is a pattern this niche is full of. There is no "Senior Audio Engineer, 15 years experience" here, because there isn't one, and we'd rather have a boring true byline than an impressive false one.

What we actually are: users of this software on our own machines, who got tired of stale advice and started writing down what we verified.

How we check things

Our house rule is that a claim about software is only as good as the day it was checked, so every fragile fact on this site carries a date and a source. Specifically:

  • Versions and release dates come from the project's own repository or release feed — not from a review site, not from a package listing that might lag.
  • Packaging claims come from the distributions' own databases. When we couldn't check a distribution, we write "not verified" rather than assuming.
  • Behaviour claims come from project documentation and from using the software. Where we haven't tested a specific combination — a distribution we don't run, a device we don't own — we say so instead of generalising.
  • Prices and plan limits for the one commercial product we mention are read off the vendor's own pages and stamped with the date, because they change without notice.
  • Rankings are computed, never typed. The scoring script ships with the site, its weights are documented in the script itself, and we re-run it rather than editing numbers in HTML.

What we refuse to do

No invented benchmarks. We do not publish signal-to-noise measurements from "our lab", because there is no lab, and in a niche where four of the five tools run the same model, precise-sounding quality numbers would be theatre. No fabricated test panels or invented listener studies. No hardware recommendations — this site is about software and settings, and audio gear is where most of the money and most of the nonsense in this field lives. And no reviews of paid products as products; we cover exactly one commercial tool, and only in the context of it not running on Linux.

On money

Almost everything we recommend is free software with no affiliate programme in existence. One product on the entire site can earn us a commission, its link is labelled everywhere it appears, and it has no Linux desktop application. We wrote out the full accounting, including the ways affiliate incentives usually bend this kind of content, in Free as in "nobody pays us".

If we're wrong

We will be, eventually — that's the nature of writing about moving software. The most likely failure mode is a date going stale: a project ships a release, a package lands in a new distribution, a price changes, and a page here quietly becomes last year's truth. If you spot one, the fastest fix is to check the source we cited and the date we cited it on; if the world has moved, the page is wrong and we'd want it corrected. Corrections get made in place, with the verification date updated, rather than quietly patched to look as though the page was always right.

Start anywhere: the stack guide if you want the whole picture, EasyEffects if you just want your fan to stop being on the call.