SnailText
EN
Pricing How it works Blog About Download for Mac For Windows

Dictation in GitHub

Voice dictation in GitHub for issues, pull request reviews, and comments

Issues and PR reviews get written badly because nobody wants to type a full description. SnailText lets you speak them, cleans the text up locally, and drops it into any GitHub field in your browser.

Download for Mac See pricing

Mac & Windows · Works in any field · Nothing uploaded

AI dictation in GitHub, on your device

SnailText is AI dictation: a speech model turns your voice into text, then a second model cleans it up, dropping filler, fixing punctuation, and matching the style your work needs. Both models run on your own machine, so nothing you say in GitHub is uploaded for transcription. Press the hotkey, speak, and the finished text lands at your cursor.

Does GitHub have built-in dictation?

The GitHub website has no built-in dictation for issues, pull requests, or comments. There is voice input in the GitHub Copilot CLI, but that is the terminal tool, not github.com. To dictate into a GitHub issue or a PR review in your browser, you need a system-wide tool. SnailText injects text at the OS level, so it works in every GitHub text field.

Speech to text in GitHub: how it works

SnailText does not plug into GitHub directly. It runs system-wide: a global hotkey (Option+Space (Mac) / Ctrl+Space (Windows), customizable) starts recording, the local speech-to-text engine (Whisper or Parakeet) transcribes what you said, and the text is pasted at your cursor through the system clipboard, the same way it lands when you type. That means it works in every GitHub field, plus every other app on your Mac or Windows machine, with no extension or integration to set up.

That matters in GitHub specifically: a good issue needs reproduction steps, context, and expected behavior, which is a lot of typing, so issues end up thin. Speech to text removes that bottleneck without changing how you work in GitHub.

Voice to text in GitHub: what to dictate

Voice typing in GitHub is not limited to one box. Press the hotkey anywhere a cursor blinks and your speech becomes text, so the things you would normally type out by hand become things you just say.

  • · Issue descriptions with reproduction steps and expected behavior
  • · Pull request descriptions and review comments
  • · Replies on threads and discussions
  • · Release notes and README sections

Where typing slows you down in GitHub

  • · A good issue needs reproduction steps, context, and expected behavior, which is a lot of typing, so issues end up thin.
  • · PR reviews are where careful written feedback matters most, and typing it slows the review down.
  • · GitHub has no native dictation on the website, so without a system-wide tool every field is keyboard-only.

Example dictations for GitHub

Bug report

"Steps to reproduce: open the settings page, switch the model while a recording is running, and the app freezes. Expected: the switch should queue until the recording stops."

PR description

"This change extracts the segment merging into its own function and adds a guard so it no longer drops the first chunk on long recordings. Tests added for both paths."

Review comment

"This looks right, but the error is swallowed here. Can we log it with the existing tracer so a failure is at least visible in the logs?"

GitHub voice dictation FAQ

Does GitHub have built-in voice dictation?
No. The GitHub website has no native dictation for issues, pull requests, or comments. Voice input exists in the GitHub Copilot CLI, but that is the terminal tool, not github.com. SnailText adds dictation to every GitHub field in your browser by inserting text at the OS level on Mac and Windows.
Can I dictate GitHub issues and PR reviews?
Yes, anywhere you can type on GitHub. Click into the issue body, a PR description, or a review comment, press the hotkey, speak, and the cleaned-up text lands at your cursor. The Development profile keeps code identifiers and library names cased correctly.
Can I use voice typing in GitHub?
Yes. SnailText adds voice typing to GitHub on Mac and Windows. Press the hotkey (Option+Space (Mac) / Ctrl+Space (Windows), customizable), speak, and the text lands at your cursor in any GitHub field, the same way it works for voice to text in every other app on your machine.
Does the dictation work offline?
Yes. SnailText runs the speech model on your device, so dictation works with no internet connection after the model is downloaded. The optional AI cleanup model runs locally too.
Is my voice uploaded anywhere?
No. Audio is processed in RAM on your machine and discarded the moment the text is ready. Nothing is sent to a server for transcription.
How much does SnailText cost?
The free tier gives you unlimited local speech-to-text with no account. Pro adds larger models and the on-device AI cleanup, $7.49/mo or $89/yr for up to 3 devices.

Dictation in other apps

Start dictating in GitHub

Free local speech-to-text, no account needed. Works in GitHub and every other app on Mac and Windows.

Download for Mac See pricing