Artificial Intelligence of YouTube is now able to describe sound effects

YouTube has the feature of captioning (video content in writing) automatically for videos on the platform since 2009. The feature has been continuously improved over time and has recently reached levels comparable to the level of human error in some languages.

Now YouTube is expanding the feature to include captions for sound effects. The video service uses the machine to detect sound effects in videos and add captions (applause, music, and laughter) to millions of videos.


These three categories have been repeatedly described manually. YouTube says it is only in the early stages of making improvements to this feature to help its deaf and hearing impaired users base.

The company says that captions such as ringing, barking and roads are in the next line and will be listed in the future, but require more time because of the lack of their databases to train artificial neural networks, which are the basic process in machine learning methods.