YouTube has just introduced the ability to provide automatic captions on all user videos.
Captions will be automatically generated through the voice recognition algorithm that is used on Google Voice. The technology is known as Automatic Speech Recognition (ASR) technology and can be pretty accurate for some cases.
In many cases, though, the automatic captions result in some rather funny output, even for videos by Google’s own employees. This could lead to some very confused users if they relied solely on automated captions!
In order to use automated captions, a user has must click on the upward arrow button on the bottom-right corner of the video control bar and click on ‘Transcribe Audio’ from the captions (CC) menu.
It is also be possible to translate the captions into several languages.
While this will greatly improve YouTube’s accessibility and benefit people who have difficulties hearing, the translation facility will also help people who do not understand the original language used in the video.
Another feature that is being simultaneously introduced is auto-timing for subtitles. If all the words used in the video are provided in a text file, Google will automatically associate the timing of each caption alongside the video timeline.
The new captioning features will therefore eventually allow users to search for a video based on keywords used in the video script.
The automatic caption feature is already available on most educational and Google-owned channels. Automatic timing is now available on all newly uploaded videos on YouTube.