YouTube Expressive Captions use AI to capture emotion, gasps, and sighs in every video

YouTube Adds Expressive Captions So You Can Feel Every Gasp, Sigh, and Shout

Android's Expressive Captions are now rolling out for YouTube videos on all devices. While traditional captions translate everything being said on screen, they don't capture the emotion behind those words. With Expressive Captions, YouTube aims to solve that problem. These captions use "AI to communicate things like tone, volume, environmental cues, and human noises from the audio playing on a device," YouTube says. Viewers will notice the difference when they see all caps for when someone's screaming with excitement or bracketed descriptions for sounds like sighs, gasps, and other background noises. Captions were originally meant for people who were deaf or hard of hearing. With the advent of streaming, however, they became essential for users who wanted to consume content in a foreign language. Expressive Captions can help both sets of users. Google originally rolled out the feature as part of Live Captions for Android phones last year, but it is now expanding it to YouTube videos in English on any device. At launch, though, it will only be available for videos uploaded after October. "We'll continue rolling this feature out on more videos," YouTube added. The video-streaming platform also launched its first-ever year-end recap this week. Like Spotify Wrapped, YouTube Recap will provide you with insights into your viewing habits for this year. It will highlight your top interests, most-visited channels, and the evolution of your preferences over the past year. YouTube Music's version of the same for songs was released last week. Google also rolled out a bunch of Android 16 updates this week.

No audio? No problem. YouTube's new Expressive Captions carry the emotion for you

Captions can now show tone, emphasis, and ambient cues like sighs, applause, or gasps. Standard captions have always been functional, but rarely fun. They get the job done if you're watching on mute or have hearing accessibility needs, but they've historically stripped the video of its emotion and energy. Well, that's officially changing, thanks to Google's latest accessibility updates. YouTube is rolling out Expressive Captions, and it's arguably one of the most interesting updates to hit the platform in a while. This feature uses AI alongside regular transcription to reveal the meaning behind the words. It's now available for English-language videos uploaded after October 2025, and YouTube plans to add it to more videos soon. Now, instead of plain text, viewers see captions that carry tone, emphasis, emotion, and even background cues like sighs, applause, or a gasp. If someone sighs or gasps, the caption will say so. You'll see tags like "[joy]" or "[sadness]" before the text, so you won't miss sarcasm or heartbreak if you can't hear the audio. This upgrade didn't appear out of nowhere. Expressive Captions first launched on Android as an evolution of Live Caption, picking up where the basic word-for-word system left off. Android's version didn't only capture what was said, but also how it was said. If someone whispered, the caption reflected that. If they yelled, you saw that intensity. If someone dragged out a word for dramatic effect, the text stretched with them. The new feature is part of Google's broader winter accessibility upgrades, which include other Android OS enhancements such as Fast Pair support for hearing aids, easier voice dictation with TalkBack, and more.