Applying our AI research to enhance the YouTube experience<\/strong><\/h4>\nHelping enrich people\u2019s lives with our research, we\u2019ve partnered with businesses across Alphabet to apply our technology towards improving the products and services used by billions of people every day.\u00a0<\/p>\n
One of our key partners is YouTube, who are on a mission to give everyone a voice and show them the world.\u00a0<\/p>\n
Working together with YouTube\u2019s product and engineering teams, we\u2019ve helped optimise the decision-making processes that increase safety, decrease latency, and enhance the viewer, creator, and advertiser experience for all. <\/p>\n\n![\"\"\/](\"https:\/\/assets-global.website-files.com\/621e749a546b7592125f38ed\/62d002be7d88680e39732a2a_YouTube_Blog__03.png\")
<\/div>\n<\/figure>\nOptimising video compression\u00a0<\/h4>\n
With video surging during the COVID-19 pandemic, and the total amount of internet traffic expected to grow in the future, video compression is an increasingly important problem.<\/p>\n
Working together with YouTube, we explored the potential for our AI model, MuZero<\/a>, to improve the VP9 codec, a coding format that helps compress and transmit video over the internet. Then we applied MuZero to some of YouTube\u2019s live traffic.<\/p>\nSince launching to production on a portion of YouTube\u2019s live traffic, we\u2019ve demonstrated an average 4% bitrate reduction across a large, diverse set of videos. Bitrate helps determine the computing ability and bandwidth needed to play and store videos \u2013 impacting everything from how long a video takes to load to its resolution, buffering, and data usage.\u00a0<\/p>\n
By improving the VP9 codec on YouTube, we\u2019ve helped reduce internet traffic, data usage, and time needed for loading videos. And through optimising video compression, millions of people around the world are able to watch more videos while using less data.<\/p>\n\n![\"\"\/](\"https:\/\/assets-global.website-files.com\/621e749a546b7592125f38ed\/62d0030b99162b7b1380f5f0_YouTube_Blog__02.png\")
<\/div>\n<\/figure>\nProtecting brand safety for creators and advertisers<\/h4>\n
Since 2018, we\u2019ve collaborated with YouTube to better educate creators on what types of videos can earn revenue from ads and make sure ads appear alongside content that follows YouTube\u2019s advertiser friendly guidelines.<\/p>\n
Together with the YouTube team, we developed a label quality model (LQM) that helps label videos with greater precision, according to YouTube\u2019s ad friendly guidelines. The model improved the accuracy of advertisements running on videos in line with YouTube\u2019s ad friendly policies.<\/p>\n
Through improving how videos are identified and classified, we\u2019ve enhanced trust in the platform for viewers, creators, and advertisers alike.<\/p>\n\n![\"\"\/](\"https:\/\/assets-global.website-files.com\/621e749a546b7592125f38ed\/62d0035179432f01d467fc1d_YouTube_Blog__04.png\")
<\/div>\n<\/figure>\nImproving AutoChapters<\/h4>\n
In recent years, creators started adding chapters to their videos to make it easier for their audience to find the content they were looking for, but this manual process can be slow and laborious.\u00a0<\/p>\n
To improve the creator and viewer experience, we collaborated with the YouTube Search team and developed an AI system that can automatically process video transcripts, audio and visual features and suggest chapter segments and titles for YouTube creators.<\/p>\n
<\/div>\n<\/figure>\nOptimising video compression\u00a0<\/h4>\n
With video surging during the COVID-19 pandemic, and the total amount of internet traffic expected to grow in the future, video compression is an increasingly important problem.<\/p>\n
Working together with YouTube, we explored the potential for our AI model, MuZero<\/a>, to improve the VP9 codec, a coding format that helps compress and transmit video over the internet. Then we applied MuZero to some of YouTube\u2019s live traffic.<\/p>\n Since launching to production on a portion of YouTube\u2019s live traffic, we\u2019ve demonstrated an average 4% bitrate reduction across a large, diverse set of videos. Bitrate helps determine the computing ability and bandwidth needed to play and store videos \u2013 impacting everything from how long a video takes to load to its resolution, buffering, and data usage.\u00a0<\/p>\n By improving the VP9 codec on YouTube, we\u2019ve helped reduce internet traffic, data usage, and time needed for loading videos. And through optimising video compression, millions of people around the world are able to watch more videos while using less data.<\/p>\n Since 2018, we\u2019ve collaborated with YouTube to better educate creators on what types of videos can earn revenue from ads and make sure ads appear alongside content that follows YouTube\u2019s advertiser friendly guidelines.<\/p>\n Together with the YouTube team, we developed a label quality model (LQM) that helps label videos with greater precision, according to YouTube\u2019s ad friendly guidelines. The model improved the accuracy of advertisements running on videos in line with YouTube\u2019s ad friendly policies.<\/p>\n Through improving how videos are identified and classified, we\u2019ve enhanced trust in the platform for viewers, creators, and advertisers alike.<\/p>\n In recent years, creators started adding chapters to their videos to make it easier for their audience to find the content they were looking for, but this manual process can be slow and laborious.\u00a0<\/p>\n To improve the creator and viewer experience, we collaborated with the YouTube Search team and developed an AI system that can automatically process video transcripts, audio and visual features and suggest chapter segments and titles for YouTube creators.<\/p>\n<\/div>\n<\/figure>\nProtecting brand safety for creators and advertisers<\/h4>\n
<\/div>\n<\/figure>\nImproving AutoChapters<\/h4>\n