Hungarian audio and video to text

Regional and cross-border Hungarian

Standard Hungarian is the easiest case. Palóc, Székely and Csángó speech, and the Hungarian of communities in Slovakia, Serbia, Ukraine and Transylvania, each have their own words and sounds, and may give more mistakes. Young speakers' slang and English words in conversation are also tougher than formal talk.

Long words, vowels and spelling

Hungarian endings follow vowel harmony and stack up, as in házaimban ('in my houses'), so a single wrong ending changes the meaning. Letters like ö, ü, ő and ű matter, and the long and short forms are different letters. Words with ly and j sound the same, so spelling slips can happen. Family names come before given names in Hungarian.

What Hungarian speakers use it for

University lectures, work meetings, interviews, podcasts and YouTube videos. Hungarian has few speakers worldwide, so tools that handle it are fewer, and an editable, searchable transcript saves hours of manual typing. Upload one recording at a time, check the names and figures first, and then read the rest while listening, which is quicker than typing it from scratch.

Which model works best for Hungarian

Hungarian is supported with a good amount of training material, but less than the largest languages, so the small model makes more slips than it does for English or Spanish, especially on rare words, names and fast or noisy speech. Expect a draft that needs a proper read-through.

On tryscriptorium.com the no-account preview and regular accounts run the small model (the larger medium and large-v3 models run when you self-host Scriptorium). In an account, pick Small in the model menu rather than Auto: Auto prefers the fastest, least accurate model for long files or a busy queue.

The small model manages clear Hungarian, but long inflected words, names and rare vocabulary are where it slips, and it can pick a neighbouring ending. Read with the audio playing, and for important recordings consider self-hosting with large-v3.

Subtitles and captions in Hungarian

Hungarian words are long, so a subtitle line fits fewer words. Keep two lines of around 40 characters and break between words, never inside one. Choose formal or informal address (maga or te) consistently with what was said. SRT and VTT are UTF-8, so ő and ű display properly.

The uploader is set to Hungarian

The upload box on this page transcribes as Hungarian, so the recogniser does not have to guess the language from the first seconds. If the recording is in another language or switches languages, use Audio to text, which detects the language automatically. Automatic transcription is a draft and no accuracy figure is promised: check names, numbers and anything you will rely on.