izwi

Voice AI that runs on your own machine: speech-to-text, text-to-speech, voice cloning, diarization, and voice chat behind an OpenAI-compatible API. Desktop app, web UI, CLI, or self-hosted server.

Runs with mockssource: GitHubhomepageRustApache-2.0commit 7804eee5a944

Rust, Apache-2.0 licensed. The project labels itself: asr, desktop app, local ai, local first, on device, openai compatible api, realtime and rust.

izwi runs, with stand-ins for the services it depends on. An Argusic agent installed it in 10 minutes and hit 3 errors on a clean machine with no GPU, and the whole session was recorded.

Izwi CLI 0.1.0-beta-18 and HTTP API server compile and run on linux-x86_64 with CPU backend; all 3169+ unit tests pass across 7 workspace crates; server serves OpenAI-compatible /v1/ endpoints on port 8080 with proper health, metrics, model listing, OpenAPI docs, and web UI; desktop Tauri shell cannot build due to missing system GUI libraries.

What the agent ended up with on a clean machine, in its own words. How this is measured

Measured by Argusic on a fresh machine every time. Every number links to its evidence. Argusic Score 52.7 of 100 (the mean of this project's run scores).

At a glance

verdict
Runs with mocks
Argusic Score
52.7 / 100
cost of the verifying run
$0.14 (measured)
recorded runs
2
last tested
stars
480
forks
48
open issues
28
watchers
6
size
59 MB
created
last push

Subject data from GitHub, linked at the top of this page, refreshed . Test data by Argusic (CC BY 4.0); every number links to a run page with the full log, the recording, and their sha256 hashes.

What was measured

From the run that decided this verdict. Every line is a recorded value, not an estimate.

Time lost to each failure while testing izwi, drawn from the recorded run
Every bar is a failure the agent recorded, with the time it cost. Drawn from this run, not from anywhere else.

What went wrong, in the agent's own words

The agent's own words, worst first, never rewritten into advice.

  • izwi-core test linker SIGSEGV (stack overflow)2 minutes
  • Rust toolchain not present1 minute
  • izwi-desktop (Tauri) build fails: glib-sys, gtk3, webkit2gtk system libraries missing
Install time
10 minutes
Cold machine to finish
30 minutes
Errors hit and fixed
3 hit, 2 fixed with no human help
How the result was proved
cargo test (7 non-desktop crates) passed 3169+ tests with 0 failures; izwi-server answered 200 on /livez, /readyz, /v1/metrics, /v1/models, /, /openapi.json, /docs; CLI version --full reports v0.1.0-beta-18 CPU linux-x86_64
Model tokens used
318,740
Exact commit tested
7804eee5a944
Container image
sha256:cdd920bce783

What was not measured

The other half, so you can see exactly how far this evidence goes. A gap is printed as a gap and never filled in.

Peak GPU memory
this ran on a CPU machine, so there was no GPU to sample
Behaviour over a long run
each attempt uses the smallest documented settings and a fixed time budget, so this measures whether it runs, not how it holds up under load
Behaviour against real services
this run was verified against stand-in services only

The end of the recording

The last lines of the terminal session that decided this verdict, exactly as they were recorded.

3. izwi-desktop glib/gtk3/webkit2gtk missing → system package gap (unfixable)
DONE
echo "todo updated"' in /work/repo
 succeeded in 0ms:
todo updated
All verified. Here's the final result.
**ARGUSIC_RESULT:** {"install_succeeded": true, "launch_succeeded": true, "install_minutes": 10, "e…
2026-10-22T06:35:02.082515Z ERROR codex_core::session: failed to record rollout items: thread 01a16…
tokens used
318,740
All verified. Here's the final result.
**ARGUSIC_RESULT:** {"install_succeeded": true, "launch_succeeded": true, "install_minutes": 10, "e…

Replay the whole session, every command from a clean machine to this point.

How it was tested

One agent, one container, no human help. It installs the project the way its own documentation says to, runs it, and fixes what breaks. Everything below is recorded as it happened: the terminal session, the log and the exact commit. The full method.

Strengths and limits

Measured facts, not opinions. How this is written.

What went well

  • Licensed Apache-2.0, as reported by its host.
  • Measured 2 times, so the result is not a one-off.

What did not, or is not known

  • Only reached a running state against stand-in services, so nothing here proves it works against the real ones.
  • Took 10 minutes to install, slower than the median of the 10 comparable projects Argusic has measured.
  • Hit 3 errors during setup, 1 of which were not solved.

Of the 11 asr projects Argusic has installed and timed, izwi was the 7th fastest to reach a running state, and 8 of 11 reached one at all.

Also tested, in the same area

Every one of these was installed and run by Argusic on a clean machine. Nothing appears here that was not tested.

Run history

tested byattemptoutcomescorecost USDwhen
Argusic Runner2/3Runs with mocks85.330.14
Argusic Runner1/3Did not run20.000.11

Topics (from GitHub)

asrdesktop-applocal-ailocal-firston-deviceopenai-compatible-apirealtimerustself-hosted-aispeaker-diarizationspeech-recognitionspeech-synthesisspeech-to-texttauritext-to-speechttsvoice-agentsvoice-aivoice-chatvoice-cloning

Embed the badge

Markdown for the project README. It links back here; terms on the terms page.

[![Tested by Argusic](https://argusic.com/badge/izwi.svg)](https://argusic.com/subject/izwi)

Questions

Does izwi run?
izwi runs, with mocks standing in for external services. Argusic installed and launched it on a clean machine in 10 minutes, hitting 3 errors on the way, and recorded the session.
How did Argusic test izwi?
On a fresh, disposable machine, with every command recorded and the repository pinned at commit 7804eee5a944. 2 attempts are recorded, and the full method is on the methodology page.
What did it cost to test izwi?
The run that produced this verdict cost $0.14: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
How long does izwi take to install?
10 minutes in the run that decided this verdict, from a clean machine to a running state. That is one measured attempt, not an average.
Does izwi need a GPU?
Not to start. Argusic reached a running state on a machine with no GPU.
What goes wrong when installing izwi?
3 things broke in the recorded run, and 2 were fixed without human help. Each one, and the time it cost, is listed on this page.
How does izwi compare with the alternatives?
Of the 11 asr projects Argusic has installed and timed, izwi was the 7th fastest to reach a running state, and 8 of 11 reached one at all.
Where is the evidence for izwi?
All 2 recorded runs are on this page, each linking to its full log and terminal recording, stored with a sha256 fingerprint so it cannot be quietly altered.

Discussion