Translate subtitles
to Russian

How it works for Russian

Upload an English .srt or .vtt file, or a season as a zip. Cues travel to the model in batches of 32 so a sentence that spans two cues is translated as one thought. Every Russian cue then passes thirteen structural checks; whatever fails goes back with the exact complaint attached, and an editor pass sweeps the whole file for register and idiom. You download the subtitles and a quality report, and the upload is deleted.

What the Russian profile checks

• Sound labels are verbal nouns, the convention Russian subtitling uses, so no table entry has an aspect or an agreement to get wrong.
• «…» quotation marks by instruction; 53 Russian function words required, 30 known intrusions blocked.
• Scored 39.72 chrF++ against a human translation. That number is low partly because chrF++ counts word n-grams and Russian inflection turns near misses into misses; it is a baseline for Russian, not a ranking against other languages.
• A mined glossary was tested for Russian and made it worse (-0.91 over three runs), so nothing is forced onto the model's spelling of names.

Sound labels, from the table

Labels for the deaf and hard of hearing are resolved from a fixed Russiantable before the model sees the text, so they are never invented. Three entries, exactly as the table renders them:

[LAUGHS]

[СМЕХ]

[DOOR CLOSES]

[ЗВУК ЗАКРЫВАЮЩЕЙСЯ ДВЕРИ]

[PHONE RINGS]

[ЗВОНОК ТЕЛЕФОНА]

Is Russian measured?

Russian has a measured baseline: 39.72 chrF++ against a professional human translation, over 1,480 aligned units. That figure is what every change to the pipeline has to beat for Russian. It is not comparable to another language's figure; here is why.

Price and time

One credit translates about 20,000 characters of dialogue, roughly a 25-minute episode; a feature film is about three. Credits are $1.25 each in packs, the first episode is free, and there is no subscription. An episode takes a few minutes; you get an email when it is ready.

Reading

• How we score a machine translation against a human subtitler