Captions for videos in these languages can be created using speech recognition
YouTube is expanding support for its automatic captions feature. The technology makes it possible to create captions for any YouTube video via speech analysis. Google's speech recognition features are getting better and better and YouTube has now added six more languages to the feature.Automatic captions had already been available for videos in English, Japanese, Korean and Spanish. Now, the feature is available for German, Italian, French, Portuguese, Russian and Dutch as well – six more languages.
“Now in 10 languages, automatic captions are an important first step in the path toward high-quality captions for the 72 hours of video people upload per minute,” YouTube wrote.
The technology is by no means perfect; speech recognition has problems even in the best of conditions. Trying to analyze speech in videos is far from ideal.
However, the feature could serve as a starting point for manual captions; video creators can generate the automatic captions and then correct the errors as needed.
The same technology is used to sync transcripts which can be uploaded to YouTube to the videos they are for.