Video to text accuracy: what we measured, including the clips that went badly

Eight fixed test clips, 229 reference words, 45 words came back wrong, missing or added — a weighted word error rate of 19.7% for the model this page ships. The best bucket was 6.1% and the worst was 33.7%. Below is every bucket, the worst single clip, and how the number was produced.

Accuracy by condition

All eight clips were run through the same model that ships with this page (whisper-base, quantised) on 2026-09-18, and the output was compared word by word against a reference transcript. Every word that came back wrong, missing or added was counted as an error.

The 19.7% is not the average of the five rows. It is 45 error words divided by 229 reference words, so the longer buckets weigh more. Adding the per-bucket percentages and dividing by five gives a different and wrong number.

The worst clip, on its own

One clip accounts for most of the damage. E1-real-noise-heavy is 29.4 seconds of real speech with heavy background noise, 71 reference words, and it came back with 28 of those 71 words wrong — 39.4%. The other heavy-noise clip in the same bucket (E2, 18 words) came back at 11.1%. Same condition, same model, 28 points apart.

That spread is the honest part of this page. A single averaged accuracy figure hides it: if we had only run E2, we would have reported a heavy-noise number four times better than the one E1 actually produced.

What this means if you are about to use the tool

What we did not test

Where the raw data is

Every number on this page comes from two files: wer-by-condition.csv (the five buckets plus the total) and wer-by-sample.csv (all eight clips individually, with the per-clip error counts). Both are published, along with the measurement scripts and the reference transcripts, so the arithmetic above can be checked or re-run.

Run it on your own file

The tool is on the front page of this site. Drop in an audio or video file you already have, pick the spoken language, and read the transcript. Nothing is uploaded — the model runs in your browser tab and the file never leaves your device.

Transcribe a file