xcodec
AAAI 2025: Codec Does Matter: Exploring the Semantic Shortcoming of Codec for Audio Language Model
Runssource: GitHubPythonMITcommit 4a737f2ee2cd
Python, MIT licensed. The project labels itself: audio, audio codec, codec, gpt, language model, music, self supervised learning and semantic.
xcodec runs. An Argusic agent installed it in 5 minutes and hit 3 errors and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.
X-Codec model builds, imports, runs encode/decode on real audio across all bandwidths, and produces valid WAV 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.
- torchcodec requires CUDA shared lib not present on CPU torch2 minutes
- scipy.signal.kaiser moved to scipy.signal.windows in scipy>=1.121 minute
- scipy==1.10.1 not available for Python 3.12
- Install time
- 5 minutes
- Cold machine to finish
- 8 minutes
- Errors hit and fixed
- 3 hit, 3 fixed with no human help
- How the result was proved
- python3 encode/decode round trip: model.encode(wav, target_bw=0.5/1.0/2.0/4.0) produced codes, model.decode(codes) produced valid 16kHz mono WAV verified by ffprobe and ffmpeg -v error -f null - across all bandwidths
- Model tokens used
- 215,331
- Exact commit tested
- 4a737f2ee2cd
- 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 1, 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.
Here's the complete report: **Install**: Set up a Python venv, installed PyTorch CPU, torchaudio, soundfile, transformers (for … **Errors encountered**: 1. **`scipy==1.10.1` not available for Python 3.12** — Fixed by removing the version pin in `requir… 2. **`from scipy.signal import kaiser` moved** — Fixed by changing to `from scipy.signal.windows im… 3. **`torchcodec` requires CUDA shared lib** — Installed torchcodec but it depends on libcuda.so wh… **Verification**: - The project builds and imports correctly. - The X-Codec model (`SoundStream` with semantic teacher) initializes, loads the Hubert model from … - The non-semantic `SoundStream` (base DAC variant) also works. - The pre-existing `test_audio_reconstruction/speech_en_nq_1.wav` is a valid WAV file (PCM s16le, 1… - The bundled descript-audio-codec tests fail only due to CUDA requirement; the DAC model works on …
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 5 minutes, faster than the median of the 17 comparable projects Argusic has measured.
- Recovered from all 3 errors 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.
What did not, or is not known
- Measured once so far. A single attempt shows it can run, not that it runs reliably.
Of the 18 audio projects Argusic has installed and timed, xcodec was the 7th fastest to reach a running state, and 15 of 18 reached one at all.
What it is a reasonable choice for
- Trying it on a laptop or a small server: it reached a running state without a GPU.
- A quick evaluation: it was running 5 minutes after the clone.
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)
audioaudio-codeccodecgptlanguage-modelmusicself-supervised-learningsemanticsoundspeechspeech-language-modeltext-to-musictext-to-soundtext-to-speechtokenizervall-e
Embed the badge
Markdown for the project README. It links back here; terms on the terms page.
[](https://argusic.com/subject/xcodec)Questions
- Does xcodec run?
- Yes. xcodec runs. Argusic installed and launched it on a clean machine in 5 minutes, hitting 3 errors on the way, and recorded the session.
- How did Argusic test xcodec?
- On a fresh, disposable machine, with every command recorded and the repository pinned at commit 4a737f2ee2cd. One attempt is recorded, and the full method is on the methodology page.
- What did it cost to test xcodec?
- The run that produced this verdict cost $0.03: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
- How long does xcodec take to install?
- 5 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 xcodec need a GPU?
- Not to start. Argusic reached a running state on a machine with no GPU.
- What goes wrong when installing xcodec?
- 3 things broke in the recorded run, and 3 were fixed without human help. Each one, and the time it cost, is listed on this page.
- How does xcodec compare with the alternatives?
- Of the 18 audio projects Argusic has installed and timed, xcodec was the 7th fastest to reach a running state, and 15 of 18 reached one at all.
- Where is the evidence for xcodec?
- The recorded run is on this page, each linking to its full log and terminal recording, stored with a sha256 fingerprint so it cannot be quietly altered.