Skip to content

[Suggestion] Chrome browser extension for live tab audio transcription (TypeScript) #265

@deepgram-robot

Description

@deepgram-robot

What to build

A Chrome browser extension that captures audio from the active browser tab (meetings, videos, podcasts) and provides real-time transcription using Deepgram's streaming STT API, displayed in a floating sidebar overlay.

Why this matters

Developers and end-users frequently need to transcribe audio playing in their browser — video calls (Google Meet, Zoom web), YouTube videos, podcasts, webinars. A Chrome extension is the most natural delivery mechanism for this use case, and building one with Deepgram showcases the streaming API's low-latency capabilities in a real-world, highly visible context. This example also demonstrates key browser patterns: tab audio capture via chrome.tabCapture, WebSocket streaming from a service worker, and secure API key handling.

Suggested scope

  • Language: TypeScript
  • Platform: Chrome Extension (Manifest V3)
  • Deepgram APIs: STT (Nova-3 streaming WebSocket)
  • Features:
    • Capture audio from active tab using chrome.tabCapture API
    • Stream audio to Deepgram via WebSocket from service worker
    • Display real-time transcription in a floating sidebar (content script)
    • Toggle transcription on/off via extension popup or keyboard shortcut
    • Copy transcript to clipboard
    • Basic speaker diarization display
  • Includes: Manifest V3 configuration, service worker for WebSocket, content script for UI overlay
  • Complexity: Intermediate-Advanced

Acceptance criteria

  • Installable as unpacked Chrome extension (load in developer mode)
  • README explains Chrome extension architecture and setup steps
  • Uses current SDK version or direct WebSocket API
  • Captures and transcribes tab audio in real-time with < 500ms visible latency
  • Sidebar overlay does not interfere with page content
  • Handles tab switching and extension lifecycle gracefully
  • Includes secure API key configuration (options page or environment variable)

Raised by the DX intelligence system.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions