Simple STT

Speech-to-text, simplified. Live transcription, read-aloud, voice commands, and a clean transcript workspace.

Open The Hosted App

Simple STT is a speech-to-text browser app and extension with live transcription, read-aloud, configurable voice commands, and optional active-field insertion.

What It Includes

• a full-page transcript workspace
• a compact popup launcher
• a settings surface for language, voice, hotkey, and command phrases
• a hosted web app at simple-stt.github.io
• optional writing of final dictated text into the currently focused editable field in the active tab

Features

• One live transcript surface with interim and final speech merged into the same textarea
• Start and stop transcription from the main app
• Read-aloud with play, pause, resume, and restart controls
• Language selection and read-aloud voice selection
• Copy, cut, and clear transcript actions
• Configurable spoken command phrases for line and paragraph breaks
• Configurable transcription toggle hotkey
• Optional active-field writing for normal web pages
• Guards against restricted browser and internal pages
• Snackbar feedback for clipboard actions, settings changes, and important errors

Open The App

Hosted web app

Visit simple-stt.github.io

Extension app

• Click the extension toolbar icon, then choose Open App
• Open settings from the popup gear or the app gear

The extension deduplicates its own app and settings tabs, so opening them again focuses the existing tab instead of creating a new copy.

Install The Extension

1. Open chrome://extensions in Google Chrome
2. Enable Developer mode
3. Click Load unpacked
4. Select this repository's ext/ directory

Chrome only:

• This project is built and tested for Google Chrome
• Brave and other Chromium browsers are not supported targets for this repo

How Transcription Works

• The main app page is the primary workspace
• Interim speech stays visible in the same textarea while speaking
• Final speech is committed into the transcript
• The transcript grows until a visual cap, then scrolls internally
• The textarea auto-focuses on load and regains focus after the main transcript actions

Read Aloud

Simple STT can read the transcript back using the browser speech-synthesis engine.

• If text is selected, it reads the selection
• If no text is selected, it can read from the cursor position
• Restart jumps back to the beginning and starts again
• Starting read-aloud stops active transcription so the app does not transcribe its own output

Active-Field Writing

When Write to active field is enabled in the main app, final dictated text is also inserted into the currently focused editable field in the active tab.

Notes:

• This only applies to normal editable pages and fields
• Restricted browser and internal pages are skipped quietly
• Turning the toggle off keeps transcription local to Simple STT
• The setting is persisted through the shared settings layer

Spoken Commands

Simple STT replaces configured spoken phrases after recognition finalizes.

Default phrases:

• carriage return => newline
• double carriage return => blank line

You can change these in Settings.

Keyboard Shortcuts

Default transcription toggle:

• Alt+Shift+R

Transcript actions in the app page:

• Ctrl/Cmd+A selects the transcript
• Ctrl/Cmd+C copies the transcript
• Ctrl/Cmd+X cuts the transcript

The transcription toggle hotkey is configurable in Settings.

Expected hotkey format:

• Alt+Shift+R
• Cmd+Shift+R
• Ctrl+Alt+K

If the saved value is blank or invalid, Simple STT falls back to the default.

Settings

Settings currently support:

• line break phrase
• paragraph break phrase
• transcription toggle hotkey
• language
• read-aloud voice

The live Write to active field mode toggle stays on the main app because it is intended as a working-mode control rather than a static preference.

Permissions

• activeTab: lets the extension interact with the current tab when needed
• scripting: used for focused-field insertion into editable pages
• storage: stores settings such as command phrases and the hotkey
• tabs: used for app and settings tab focusing and deduping

Limitations

• Speech recognition depends on the browser's built-in speech recognition support
• Behavior is intended for Google Chrome only
• Active-field writing will not work on restricted browser and internal pages
• Dictation quality and availability depend on the browser speech engine and microphone permissions

License

MIT. See LICENSE.