Getting Started

Subtitle Insights relies on experimental Chrome built-in AI APIs. Follow these steps to set up your environment and install the extension.

Prerequisites

• Chrome Version: You must use Chrome 138 or higher.
• Hardware: For specific hardware requirements and supported devices, please refer to the official Chrome Prompt API documentation.

1. Enable Chrome Flags

You must enable several experimental flags to allow local AI processing:

1. Open chrome://flags in your browser.
2. Search for and Enable the following flags:
   • #optimization-guide-on-device-model (Set to Enabled BypassPrefocals)
   • #prompt-api-for-gemini-nano (Set to Enabled)
   • #translation-api (Set to Enabled)
3. Relaunch Chrome.

2. Install the Extension

Since Subtitle Insights is in active development, you must install it in developer mode:

1. Download or clone the repository from GitHub.
2. Build the extension:

   npm install
   npm run build

3. Open chrome://extensions.
4. Enable Developer mode (top right toggle).
5. Click Load unpacked and select the dist/ directory from the project folder.

3. Configuration & Verification

1. Configure your Profile: Navigate to any YouTube video and click the Subtitle Insights icon in the video control bar. Open Detailed Settings to configure your profile. By default, a Japanese-to-English prompt is provided. Refresh the YouTube page after creating or selecting a profile to apply the changes.
2. AI Ready: Check the extension menu to confirm the AI status. "AI Ready" indicates the models are loaded and functional.
3. Downloading Models: If you see "Downloading Models," Chrome is currently fetching the necessary AI components. This may take several minutes.
4. Enable Captions: Make sure Closed Captions (CC) are enabled in the video player. Subtitle Insights scans these captions to provide the overlay, translations, and insights.