1.88 GitHub Copilot Release Highlights - バイリンガル字幕
It's time for us to once again dive into the ever-changing world of AI with everything
that's new for GitHub co-pilot in Visual Studio Code.
Let's rock.
In-line chat now starts as a floating control, which makes it more lightweight.
but after the first request, the control then expands and takes up more space, so you have more room for more activities.
We've also adjusted the rendering to be more consistent with other chat experience, such as the chat view or the quick chat.
When you're in a Jupyter Notebook, the kernel state, so for example, variables and available packages is now automatically included as context in inline chat.
And this lets Copilot use the current state of the Notebook so that you get more relevant completion.
If you haven't tried the commit message generation feature yet, what are you waiting for?
Check it out.
We've worked hard to improve this by including recent commit messages from the repo.
So check this out if you've never checked it out before, or you tried it once and you haven't in a while.
If you didn't know, the workspace participant can generate new projects with the slash new command.
This command now offers sample projects, which we've curated as suggestions when a suitable match is detected for the chat prompt.
You could also now tweak the result of slash new with a follow-up command.
Copilot is really good at explaining terminal errors,
so there is a new slash explain command available that is specifically optimized for explaining commands or errors.
Inline chat is now available in the terminal as a preview in this release.
If want to try this out, you'll need to enable it with the Experimental Inline Chat setting.
Inline chat in the terminal always uses the Terminal Chat participant, which is optimized for knowledge for your specific terminal.
Control runs the command in the terminal.
For the chat view,
there's a new drop-down next to the chat submit button that has more actions and you can do things here like sending the query to the workspace participant or just
starting a new chat which will clear the current chat and then submit the message.
And finally,
if you use the slash explain command without having anything selected in the active editor and your code has multiple scopes of interest like this function here,
then we'll prompt you to clarify which symbol or block scope that you want expected.
And that's it for this release.
As check out the official release notes for everything that's new in Visual Studio Code and GitHub Co-pilot.
And we'll see you back here next month.
And as always, happy smart coding.
さらなる機能をアンロック
Trancy拡張機能をインストールすると、AI字幕、AI単語定義、AI文法分析、AIスピーチなど、さらなる機能をアンロックできます。
主要なビデオプラットフォームに対応
TrancyはYouTube、Netflix、Udemy、Disney+、TED、edX、Kehan、Courseraなどのプラットフォームにバイリンガル字幕を提供するだけでなく、一般のウェブページでのAIワード/フレーズ翻訳、全文翻訳などの機能も提供します。
全プラットフォームのブラウザに対応
TrancyはiOS Safariブラウザ拡張機能を含む、全プラットフォームで使用できます。
複数の視聴モード
シアターモード、リーディングモード、ミックスモードなど、複数の視聴モードをサポートし、バイリンガル体験を提供します。
複数の練習モード
文のリスニング、スピーキングテスト、選択肢補完、書き取りなど、複数の練習方法をサポートします。
AIビデオサマリー
OpenAIを使用してビデオを要約し、キーポイントを把握します。
AI字幕
たった3〜5分でYouTubeのAI字幕を生成し、正確かつ迅速に提供します。
AI単語定義
字幕内の単語をタップするだけで定義を検索し、AIによる定義を利用できます。
AI文法分析
文を文法的に分析し、文の意味を迅速に理解し、難しい文法をマスターします。
その他のウェブ機能
Trancyはビデオのバイリンガル字幕だけでなく、ウェブページの単語翻訳や全文翻訳などの機能も提供します。