We recently updated the MLLP platform for transcription and translation in order to support the Slovenian language, both for automatic transcription of videos in Slovene, and automatic translation of videos from English into Slovene. Watch an example of Slovene automatic transcription, or visit our platform to try it yourself!
We at the MLLP are constantly updating our automatic transcription and translation systems, in order to support new languages and language pairs, as well as to improve those that are already supported. This week, we updated the MLLP platform for transcription and translation in order to support the Slovenian language, both for automatic transcription of videos in Slovene, and automatic translation of videos from English into Slovene.
As an example of our platform’s automatic transcription of a video in Slovene, you can watch this video: http://ttp.mllp.upv.es/s/cbysI (“Pozdravni nagovor in predstavitev video konference – Simon Delakorda”, by Inštitut za elektronsko participacijo INePA; original without subtitles at: http://youtu.be/Q-eDlpYa85g).
If you would also like to try it out for yourself, just visit the MLLP platform now and upload your own videos and audios! You’ll be able to see our automatic subtitles, and edit them using our easy-to-use but powerful subtitle editing interface.
With the addition of Slovene, the MLLP platform now supports the following 10 transcription languages:
- English, Spanish, Catalan, German, French, Estonian, Italian, Dutch, Portuguese, and Slovene.
And the following 14 translation pairs:
- English ↔ Spanish, Catalan, French, Italian
- English → Slovene
- Estonian, Dutch, Portuguese → English
- Spanish ↔ Catalan
Slovene is an example of an interesting challenge in Natural Language Processing: developing effective systems for languages for which there are not so many linguistic resources available (monolingual and bilingual corpora). With the state-of-the-art technologies developed at the MLLP, it is now possible to make the most of limited data and obtain results that were difficult to imagine until recently.
In the cases of Slovene and German, these are only our first public systems for these languages, so we expect significant improvements in the future as we continue their development. We will continue updating our systems in the MLLP platform for these and other languages.