How to Write Subtitles with GitHub Copilot

GitHub Copilot, the AI-powered coding assistant, is not limited to code generation. It can also help you write impactful subtitles for your videos, podcasts, or multimedia content. Whether you are working on a professional subtitling project or simply want to add subtitles to your creations, Copilot significantly speeds up the process. By leveraging its contextual completion and rephrasing capabilities, you can produce clear, synchronized subtitles tailored to your audience. This tutorial guides you step by step to make the most of GitHub Copilot for writing subtitles, from preparing your environment to exporting your final SRT or VTT files.

Prerequisites

1.Have a GitHub account with an active GitHub Copilot subscription
2.Have Visual Studio Code installed with the GitHub Copilot extension
3.Have a raw transcription or an audio/video file to subtitle
4.Know the basics of the SRT or VTT format for subtitles

Steps

Prepare Your Subtitle File in VS Code

Create a new file with the .srt or .vtt extension in Visual Studio Code. Structure the first lines with the standard format (sequence number, timecode, text). GitHub Copilot will use this initial structure to generate consistent subsequent suggestions. If starting from a raw transcription, paste it into a separate file to serve as contextual reference.

Open a .srt file and type the first two entries manually to establish the pattern: number, timecode, and text. Copilot will automatically complete the following entries.

Tip: Keep your raw transcription open in an adjacent tab. Copilot uses the context of open files to improve its suggestions.

Use Copilot Chat to Segment the Transcription

Open the Copilot Chat panel and ask it to split your raw transcription into segments suitable for subtitles. Each segment should contain 1 to 2 lines of text, with a maximum of about 42 characters per line. Copilot will respect natural speech pauses and punctuation to create logical cuts.

Split this transcription into subtitle segments. Each segment must be at most 2 lines of 42 characters. Respect natural pauses and punctuation. Format the result in SRT with estimated timecodes every 3 seconds.

Tip: Specify the language and desired register (formal, informal) in your prompt to get subtitles suited to your target audience.

Rephrase and Condense Overly Long Subtitles

Some segments will be too long to be read comfortably on screen. Select these passages and use Copilot Chat to rephrase them more concisely while preserving the original meaning. The goal is to maintain a comfortable reading time, typically between 150 and 200 words per minute displayed on screen.

Rephrase this subtitle so that it fits within a maximum of 2 lines of 42 characters while keeping the exact meaning. Use simple and direct language.

Tip: Use the shortcut Ctrl+I (or Cmd+I on Mac) to open Copilot's inline editor directly on the selection, speeding up quick corrections.

Adjust Timecodes and Synchronization

Ask Copilot to recalculate the timecodes based on the actual duration of your video. Provide it with key reference points (start, end, important pauses) and let it redistribute the timecodes proportionally. Then manually check the critical passages where synchronization must be perfect.

The video is 5 minutes 30 seconds long. Recalculate the timecodes of these SRT subtitles so they are proportionally distributed over this duration. The first subtitle starts at 00:00:01,000.

Tip: For precise synchronization, use a video player with timecode display (like VLC) in parallel and manually adjust key moments.

Check, Correct, and Export the Final File

Perform a full review with Copilot Chat's help to detect spelling, grammar, and SRT formatting errors. Also ask it to check the consistency of timecodes (no overlaps, no excessive gaps). Once validated, save your .srt file or convert it to .vtt if needed.

Check this complete SRT file. Correct spelling and grammar errors. Verify that no timecodes overlap and that the SRT format is valid. Flag subtitles that exceed 42 characters per line.

Tip: Install the 'Subtitle Editor' extension in VS Code to get a real-time preview of your subtitles and visually detect formatting issues.

Common mistakes to avoid

✗Creating subtitles that are too long, exceeding 2 lines or 42 characters per line, making reading uncomfortable for the viewer
✗Not providing enough context to Copilot (raw transcription, target language, register), resulting in generic and poorly suited suggestions
✗Blindly trusting the timecodes generated by Copilot without checking them against the actual video, causing synchronization delays
✗Forgetting to check the technical format of the SRT file (blank lines between blocks, timecode format with comma not period), preventing import into video players
✗Neglecting the final review of generated subtitles, which may contain meaning errors or awkward rephrasings

FAQ

Can GitHub Copilot generate subtitles directly from an audio or video file?

No, GitHub Copilot is a text completion tool and cannot transcribe audio. You must first obtain a text transcription (via Whisper, YouTube, or a transcription service) and then use Copilot to format, segment, and improve that transcription into professional subtitles.

What is the difference between SRT and VTT formats for subtitles?

The SRT (SubRip) format is the most universal and compatible with the majority of video players and platforms. The VTT (WebVTT) format is designed for the web and supports advanced features like positioning and CSS styling. Copilot can help you easily convert from one to the other.

How can I improve the quality of Copilot's suggestions for subtitles?

Keep several context files open in VS Code: the raw transcription, a glossary of specific terms, and an example of already formatted subtitles. The more context Copilot has, the more relevant and consistent its suggestions will be with your subtitling style.

How to use this prompt

Copy the prompt with the button above.
Paste it into ChatGPT, Claude or your favorite AI assistant.
Replace the bracketed variables with your details, then refine the result.

About Prompt Guide

Prompt Guide is a free library of 2500+ ready-to-use prompts for ChatGPT, Claude and other AIs, with guides to learn prompting and tools to build and optimize your own prompts.

Prompt library Learn prompting Prompt builder Prompt optimizer

More tutorials

How to Draft a Contract with GitHub Copilot

GitHub Copilot, Microsoft's AI assistant integrated into Visual Studio Code, is not limited to code generation. Thanks to its advanced natural language processing capabilities, it can help you draft structured legal documents like contracts. Whether you are a freelancer, entrepreneur, or project manager, Copilot allows you to quickly create a first draft of a contract while respecting essential clauses. In this tutorial, you will learn how to use GitHub Copilot to generate a professional contract step by step, customizing each section to your needs. Note: the result must always be reviewed by a legal professional before signing.

How to Draft Terms of Service with GitHub Copilot

Terms of Service (ToS) are a legal document essential for any website or application. They define the rules for using your service, protect your interests, and inform users of their rights and obligations. Drafting complete and compliant ToS can seem complex, but GitHub Copilot simplifies this process significantly. With artificial intelligence, you can generate a solid structure, cover essential clauses, and adapt the document to your specific context. This tutorial guides you step by step to create professional ToS using GitHub Copilot as a writing assistant, while keeping in mind that legal validation is still recommended.

How to Write a Blog Post with Claude

Writing a quality blog post takes time, research, and a good structure. With Claude, Anthropic's AI assistant, you can speed up

How to Write a Blog Post with GitHub Copilot

GitHub Copilot, the AI-powered coding assistant from OpenAI and GitHub, is not limited to code generation. It can also help you write blog posts directly from your code editor.

How to Write a Blog Post with Midjourney

Midjourney is an AI-powered image generation tool that can radically transform the visual quality of your blog posts. Rather than searching for generic stock photos, you can create unique illustrations perfectly tailored to your content. In this tutorial, we'll explore how to use Midjourney to produce impactful visuals for each section of your article: cover image, concept illustrations, stylized infographics, and brand graphic elements. Whether you're a blogger, web writer, or content creator, mastering Midjourney will enable you to publish professionally-looking articles without a graphic designer, while maintaining a consistent visual identity across your entire blog.

How to Write a Business Proposal with GitHub Copilot

GitHub Copilot, GitHub's AI assistant, is not limited to code generation. Thanks to its natural language understanding capabilities, it can help you

Get new prompts every week

Join our newsletter.