Catch the Translation Failures You Can Actually Prove
Paste the source transcript on one side and the translation on the other, and it compares them segment by segment. It reports what is missing, what was left in the original language, which numbers vanished, and where the subtitle timings stopped lining up. All of it in the browser, so an unreleased script stays on your machine.
This exists because machine translation fails in a specific, boring, checkable way: it quietly drops things. A file goes in with 40 cues and comes back with 31, and nothing in the process tells you. Nobody notices until a viewer does.
What It Checks, and Why Those Things
Every check here is something you can settle with evidence rather than opinion.
Missing segments. The count of translated cues against the source. If the translation is short, content was dropped, and on long files it is almost always the end that is gone. This is the single most common failure and the easiest to miss by eye, because the start of the file reads perfectly.
Lines left untranslated. Any segment whose text is identical to the source. Sometimes that is correct, a brand name or a product term that should not change. Often it means the translator skipped that chunk and passed the original through. The tool flags it and lets you judge which.
Numbers that disappeared. It pulls the figures out of each source segment and checks they survived into the translation. Prices, dates, percentages, version numbers. A sentence can read fluently in the target language and still have lost the number that was the entire point of it.
Timing drift. When both files carry cue timings, it compares them. A translated subtitle with shifted timings appears at the wrong moment, which is a worse viewing experience than a slightly clumsy sentence.
What This Cannot Do, Said Plainly
It does not judge whether the translation is any good.
Fluency, tone, register, whether a joke landed or a term of art was rendered correctly: none of that is checkable by comparing two strings across languages. A segment can pass every check here and still be wrong in a way only a speaker of the language will catch.
So treat this as the first gate, not the last. It catches the mechanical failures cheaply and at scale, which frees a human reviewer to spend their attention on meaning instead of counting cues. I would run this before sending anything to a reviewer, not instead of one.
The length comparison is the softest check for the same reason. Some language pairs genuinely expand or compress, German against English being the obvious one, so a flag there is a prompt to look, not a verdict.
Transcription Accuracy Is a Different Question
Worth separating two things that sound alike. Checking a transcript against the audio it came from, same language, is transcription accuracy, and the measure for that is word error rate. The word error rate calculator does that job: it aligns two versions of the same language and counts substitutions, deletions, and insertions.
This page is the cross-language version, and it cannot count word errors, because the words are supposed to be different. That is why it measures structure and completeness instead. Use WER for “did the transcription hear it right,” use this for “did the translation keep all of it.”
Before and After the Check
If you still need the translation itself, the video translator handles dubbing and subtitles, and the YouTube video translator works straight from a link. To produce the source transcript in the first place, use the transcript generator. And once a translation passes, the text to SRT converter turns it into a subtitle file with the timings intact.
