AI Voice Bridge v3.0 - Windows Whisper Edition
Copyright 2026 Reasonable IT Service. All rights reserved.
Website: https://www.reasonableitservice.com

Not affiliated with, endorsed by, or sponsored by OpenAI, Anthropic, DeepSeek, or Google.
ChatGPT, Claude, DeepSeek, and Gemini are trademarks of their respective owners.

OVERVIEW
AI Voice Bridge is a local voice bridge for supported AI chat websites in Google Chrome.
It adds local dictation, local readback, a floating Mic | Repeat | Stop control, and a simple voice command for sending messages.

Supported sites:
- ChatGPT: chatgpt.com / chat.openai.com
- Claude: claude.ai
- DeepSeek: chat.deepseek.com
- Gemini: gemini.google.com

WHAT IT DOES
1. Local dictation
   Click the Mic button, speak, then stop recording. Audio is sent only to the local server on the same PC, transcribed locally with Whisper, and inserted into the AI prompt box.

2. Stop and repeat local readback
   Click the red Stop button to stop current read-aloud speech. This does the same job as Esc.
   Click the green Repeat button to replay the latest captured AI response from the beginning using the local RITS/system voice path.
   Native ChatGPT/Claude read-aloud triggering is not included in this build.

3. Voice command send
   If your dictation ends with the exact phrase "send message" or "send a message", the bridge strips the command and submits the remaining prompt automatically.

Example:
You say: Explain this in one paragraph send message
The AI site receives: Explain this in one paragraph

The command only triggers when the send phrase is the final phrase. Saying something like "send message to the client" will not trigger auto-send.

CONTROLS
- Click the Mic icon: start recording
- Mic turns red while recording
- Click the Mic icon again: stop recording and transcribe
- Enter while recording: stop recording and transcribe
- Ctrl+Shift+V: start/stop dictation
- Click the red Stop button: stop current read-aloud speech
- Click the green Repeat button: replay the latest captured AI response from the beginning
- Esc inside the AI site tab: emergency stop for current read-aloud speech

FLOATING MIC | REPEAT | STOP CONTROL
The floating Mic | Repeat | Stop control uses the same stable default placement logic as the stable v3.0 build.
It docks near the supported chat composer when a safe placement is found.
If the page layout changes, the screen is too narrow, or the dock would overlap native controls, it falls back to the stable bottom-left placement.

You can move the control manually:
- Drag the middle divider/handle between Mic and Repeat/Stop to move the control
- The position is remembered per site/browser
- Double-click the middle divider/handle to reset back to the default placement

PRIVACY / SAFETY
- No cloud TTS
- No API key
- No RITS account
- No startup service
- No telemetry
- Nothing runs unless you launch it
- The server listens only on 127.0.0.1:5055
- Dictation audio is sent only from Chrome to the local server on the same computer
- The AI platform receives the final text prompt only when you submit it
- Raw microphone audio is not sent to ChatGPT, Claude, DeepSeek, or Gemini by AI Voice Bridge

LICENSE / USAGE TERMS
AI Voice Bridge is free to use, but it is not open-source.
You may install and use unmodified copies for personal or business use.
You may not copy, modify, redistribute, resell, rebrand, sublicense, publish the source code from, or create derivative products based on this software without written permission from Reasonable IT Service.
See LICENSE.txt for full terms.

IMPORTANT
Keep the AI Voice Bridge server window open while using supported AI sites.
Only one server can run at a time because the app uses local port 5055.

WINDOWS REQUIREMENTS
- Windows 10 or 11
- Google Chrome
- Python 3 (Windows installer included in the payload folder)
- Microphone permission for Chrome

INSTALL / UPDATE
1. Run Run-AI-Voice-Bridge-Windows-Installer.bat. If Python is missing, run the included payload/python-3.10.11-amd64.exe installer, then rerun the AI Voice Bridge installer.
2. The installer copies the app to: %USERPROFILE%\Documents\AI-Voice-Bridge
3. A Desktop shortcut and Start Menu shortcut named AI Voice Bridge are created automatically.
4. Launch the local server with Start-AI-Voice-Bridge.bat or the shortcut.
5. In Chrome, load the unpacked extension from:
   %USERPROFILE%\Documents\AI-Voice-Bridge\voice-bridge-extension
6. Open ChatGPT, Claude, DeepSeek, or Gemini in Chrome.
7. Use the floating Mic | Repeat | Stop control, Enter-to-stop, hotkey dictation, or final phrase auto-send.

ICON / ASSETS
- App icon files are stored in: %USERPROFILE%\Documents\AI-Voice-Bridge\assets
- Chrome extension icons are already included in the extension folder.

TROUBLESHOOTING
- If the Mic button says Failed to fetch: confirm the server is running on 127.0.0.1:5055.
- If the Mic button does not appear: reload the Chrome tab and confirm the extension is loaded.
- If Chrome blocks the microphone: check Chrome site permissions and Windows microphone privacy permissions.
- If the floating control is in a bad location: double-click the middle divider/handle to reset its position.
- If local readback does not speak: confirm the server is running and system audio is not muted.


UPGRADE NOTES
- Creates a Desktop shortcut automatically.
- Creates a Start Menu shortcut automatically.
- Includes custom app icon assets (.ico + .png) in the app folder.
- Includes Chrome extension icons in the extension package.
