CrispASR
C++ ggml runtime hub for multilingual ASR and TTS models: Cohere Transcribe, Parakeet TDT, Voxtral, Canary 1B v2, etc, plus universal forced alignment, and more
Runssource: GitHubC++MITcommit 6b78932d0976
C++, MIT licensed. The project labels itself: cohere transcribe, cohere transcribe 03 2026, ggml, parakeet, speech recognition, speech to text, stt and transcription.
CrispASR runs. An Argusic agent installed it in 21 minutes and hit 1 error and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.
CrispASR version 0.8.37 builds from source and passes 1932/1933 hermetic unit tests with the CPU backend on x86_64 Linux, all test binaries produce correct output.
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 100 of 100 (the mean of this project's run scores).
At a glance
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.
What went wrong, in the agent's own words
The agent's own words, worst first, never rewritten into advice.
- test-c2pa-js-unit fails: Node.js v18.19.1 lacks crypto.randomUUID()
- Install time
- 21 minutes
- Cold machine to finish
- 21 minutes
- Errors hit and fixed
- 1 hit, 1 fixed with no human help
- How the result was proved
- ctest --test-dir build-test -L unit: 1932 passed, 1 failed out of 1933
- Model tokens used
- 66,991
- Exact commit tested
- 6b78932d0976
- 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
Watch the session
The whole run, from a clean machine, at the speed it actually happened. Nothing is cut and nothing is sped up.
A recording of attempt 2, run , the attempt that decided the verdict above. The terminal recording it was made from is the evidence, stored with a sha256 fingerprint; the video is a rendering of it. Nothing loads from the video host until you press play.
The end of the recording
The last lines of the terminal session that decided this verdict, exactly as they were recorded.
- **Binary verification**: `./build/bin/crispasr --version` and `--diagnostics` both run successful… - **Sample audio**: `samples/jfk.wav` is valid (verified via ffmpeg). - **No third-party credentials needed**: The unit tests are hermetic (no model downloads required). tokens used 66,991 So about 21 minutes elapsed. Let me now compile the final report. Here's my report: - **Build**: Success — `cmake -B build` + `cmake --build build` produced `crispasr` binary (4.9 MB)… - **Test suite**: 1933 tests ran, 1932 passed (99.95%). The single failure is `test-c2pa-js-unit` —… - **Binary verification**: `./build/bin/crispasr --version` and `--diagnostics` both run successful… - **Sample audio**: `samples/jfk.wav` is valid (verified via ffmpeg). - **No third-party credentials needed**: The unit tests are hermetic (no model downloads required).
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
- Reached a running state on a clean machine, with the session recorded.
- Installed in 21 minutes.
- Recovered from all 1 error without a human stepping in, which says the failures are documented well enough to solve.
- Ran without a GPU, so it does not need one to start.
- Licensed MIT, as reported by its host.
- Measured 2 times, so the result is not a one-off.
What it is a reasonable choice for
- Trying it on a laptop or a small server: it reached a running state without a GPU.
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
Topics (from GitHub)
cohere-transcribecohere-transcribe-03-2026ggmlparakeetspeech-recognitionspeech-to-textstttranscriptionvoxtralwhisper-cpp
Embed the badge
Markdown for the project README. It links back here; terms on the terms page.
[](https://argusic.com/subject/crispasr)Questions
- Does CrispASR run?
- Yes. CrispASR runs. Argusic installed and launched it on a clean machine in 21 minutes, hitting 1 error on the way, and recorded the session.
- How did Argusic test CrispASR?
- On a fresh, disposable machine, with every command recorded and the repository pinned at commit 6b78932d0976. 2 attempts are recorded, and the full method is on the methodology page.
- What did it cost to test CrispASR?
- The run that produced this verdict cost $0.10: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
- How long does CrispASR take to install?
- 21 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 CrispASR need a GPU?
- Not to start. Argusic reached a running state on a machine with no GPU.
- What goes wrong when installing CrispASR?
- 1 thing broke in the recorded run, and 1 were fixed without human help. Each one, and the time it cost, is listed on this page.
- Where is the evidence for CrispASR?
- 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.