In a significant expansion beyond accessibility, Windows 11’s Live Captions feature now translates spoken audio from more than 40 languages into English in near real time—and from a subset into Simplified Chinese—directly on Copilot+ PCs. The capability, which Microsoft has been rolling out through its Copilot+ initiative, turns any app, video call, or media stream into a multilingual experience without relying on cloud processing.

What Just Happened to Live Captions?

Windows 11’s Live Captions started as an on-device subtitling tool for the hearing impaired. It could capture any audio playing through the PC—whether from a browser, a meeting app, or a local file—and display text captions without an internet connection. That core functionality remains, but Microsoft has layered real-time translation on top for devices that meet the Copilot+ hardware spec.

On qualifying hardware, Live Captions now automatically transcribes spoken audio in its original language and then translates it into English or Simplified Chinese. The translation covers over 40 source languages for English output and 27 for Chinese, according to Microsoft’s documentation. This works system-wide: the translated captions appear in the same overlay as regular captions, so they follow you across applications.

The catch? You need a Copilot+ PC—a laptop powered by a processor with a dedicated neural processing unit (NPU) that accelerates AI tasks. These include Snapdragon X-series chips, Intel Core Ultra processors, and AMD Ryzen AI systems, but the feature is still reaching devices in phases. Early availability focused on Qualcomm-powered machines; Intel and AMD support is now being tested and gradually deployed.

What Does This Mean for You?

The impact depends on how you use your PC.

For everyday users, the translation layer is a quiet superpower. Watching a foreign film on a streaming service that lacks subtitles? Live Captions will catch the dialogue and translate it on the fly. Jumping into a multilingual video call? You’ll see captions in your language without needing a third-party app. The feature works even when your device is offline, though initial setup may require downloading language packs.

For professionals and frequent collaborators, Live Captions removes a major friction point. Remote meetings with international colleagues become more manageable when you can read real-time translations of spoken remarks. Students attending lectures in a non-native language can follow along more easily. Journalists and researchers can quickly capture quotes or timestamps from interviews in other languages. However, accuracy isn’t perfect—accents, background noise, and specialized jargon can trip up the transcription and translation models. It’s best treated as a comprehension aid, not a certified transcript.

For accessibility, the feature remains a lifeline. People with hearing loss already rely on Live Captions to engage with audio content. The addition of translation means they can now also participate in multilingual environments without extra tools.

How We Got Here

Live Captions didn’t appear overnight. It began as an accessibility hackathon project at Microsoft, later evolving into a system-level feature in Windows 11. The goal was simple: let users caption any audio on their PC without relying on individual apps to support subtitles.

The feature launched in 2022 as an online-only tool, but Microsoft quickly moved to on-device processing to protect privacy and enable offline use. That shift was critical: it meant speech data never left the device, and captions were generated locally without being stored.

The Copilot+ hardware push, announced in 2024, aimed to bring AI acceleration to everyday tasks. Live Captions became a flagship user-facing example, demonstrating how an NPU can handle real-time transcription and translation with low latency. The translation feature first appeared in Insider previews for Snapdragon Copilot+ PCs, then expanded to Intel and AMD platforms as driver support matured.

What You Should Do Now

1. Turn on Live Captions
Press Windows key + Ctrl + L to toggle the feature. The first time you enable it, Windows will prompt you to download necessary speech recognition packages. If you’re on a Copilot+ PC, make sure your Windows build is up to date to get translation capabilities.

2. Position the captions for your workflow
Open the Live Captions window, click Settings > Position, and choose Above screen, Below screen, or Overlaid on screen. The docked options reserve screen space so captions don’t obscure other windows, while the overlay can be dragged anywhere.

3. Customize readability
Go to Settings > Preferences > Caption style and pick a theme, or hit Edit to tweak text size, color, and background opacity. A custom style can make captions easier to spot during busy meetings.

4. Check your hardware
Translation works only on Copilot+ PCs with an NPU. If you’re unsure, look for the Copilot+ badge on your device or check your processor specs. Even without Copilot+, you can still use standard live captions.

5. Improve accuracy
Use a good microphone close to the speaker, minimize background noise, and encourage clear turn-taking in conversations. Better audio input yields better transcription and translation results.

6. Test before relying on it
Run a trial call or play a video in your target language to gauge latency and accuracy before an important meeting. If captions lag or garble words, you may need to adjust your audio setup.

What to Watch Next

Microsoft’s phased rollout suggests more improvements are coming. Expect support for additional translation languages, lower latency as NPU drivers mature, and broader device compatibility. The Insider channels are already testing tweaks, and feedback from early adopters will shape how fast these changes reach stable builds.

In the long run, as AI accelerators become standard in laptops, real-time translation could become a baseline Windows feature rather than a premium add-on. For now, if you own a Copilot+ PC, take a few minutes to enable Live Captions and see how it fits into your daily routine. It’s one of those rare features that starts as an accessibility tool and ends up making everyone’s life a little easier—quietly, and often without an internet connection.