Videos uploaded to YouTube, Google’s video hosting service, are typically available in one language only. Google introduced subtitle options to YouTube long ago to provide these options to users who do not understand the language.
YouTube now offers automatic caption generation. This system provides access to content to individuals who would not otherwise be able to access it. However, automatic caption generation does not work well.
Despite the original language of the video not being one of the languages that YouTube users are familiar with, they may soon be able to listen to videos in languages they are familiar with. By utilizing artificial intelligence, videos may be dubbed automatically on YouTube soon through Aloud, a Google Area 120 project.
The Aloud project was first introduced in 2022 as a project within Area 120. Back then, the company’s primary objective was to create tools that would assist video producers in creating subtitles for their videos in different languages and dubbing.
A dub is a video production technique in which the language of a video is changed through the swapping of audio tracks and an attempt is made to match the new audio track with the lip movements as closely as possible.
In terms of video production, Aloud does most of the heavy lifting. At the beginning of the video, it transcribes the video, which can be reviewed and edited by the video producers. The text is translated into selected languages and the dub is produced once Aloud is satisfied with the results.
Aloud is in early access at the moment, which restricts both the technology’s accessibility and the languages it supports. However for now it is only supporting a few languages only which also include English, Spanish, and Portuguese.
An Amjad Hanif, Vice President of Creator Products, YouTube statement indicates that YouTube is currently testing the tool with some creators.
The following video can be viewed on YouTube by interested users and content producers. Although it was originally published in English, Aloud has made it available in Spanish as well.
Change the audio track to Spanish (Latin America) dubbed by selecting Settings > Audio Track > Spanish (Latin America) dubbed.
Aloud is part of YouTube’s plans. The Aloud application will also provide support for other languages in addition to mimicking the creator’s voice in the target language, improving lip-syncing, and adding more expression.
The applications of this technology are numerous. Now Content Creators at YouTube can use this technology to gain their reach of audience which was not possible in the past.
However, it is still left to be seen how flawlessly the transcription and foreign language audio creations will need more investment in terms of work from the producers to create a final good transcription that the AI may use to create the additional audio tracks for their videos.