Live Caption Translator

Real-time live captions with translation using OpenAI Realtime and Groq

Live Caption Translator adds real-time captions with translation to any webpage. Perfect for videos, meetings, or live speech. FEATURES • Tab Audio — Caption audio/video playing in the current browser tab (YouTube, podcasts, etc.) • Microphone — Caption your own voice in real time (meetings, calls, presentations) • Translation — See captions in your preferred language • OpenAI Live Translate — Stream translated transcript deltas directly with gpt-realtime-translate • Optional translated audio — Play OpenAI Live Translate audio output when available • Auto-detect — Source language can be set to auto-detect or manually selected • Choose your provider — OpenAI Realtime or Groq Whisper • Customizable overlay — Adjust text colors, background, opacity, and font size • Draggable and resizable — Move the caption panel anywhere on the page HOW IT WORKS 1. Open the Credentials tab and choose your provider (OpenAI or Groq) 2. Enter your API key for the selected provider (stored in Chrome storage for your use; not shared except with the selected API provider) 3. Choose Tab Audio or Microphone, and set your source and target languages 4. Open a regular web page (not chrome:// or new tab), then click Start Captioning 5. Captions appear in a floating overlay that you can drag and resize REQUIREMENTS • Your own API key — either OpenAI (platform.openai.com) or Groq (console.groq.com) • A regular web page must be open — captioning cannot run on chrome://, new tab, or extension pages PROVIDERS • OpenAI Live Translate — Dedicated realtime translation with gpt-realtime-translate; source transcript via gpt-realtime-whisper; optional translated audio output. • Groq — Ultra-fast Whisper transcription (whisper-large-v3 / whisper-large-v3-turbo). Translation via llama-3.3-70b-versatile. ---------------------------------------------------------------------------- Virtual Audio Devices — Windows & macOS Virtual audio devices let you route sound between apps internally. This can be useful with Live Caption Translator when you want to: - Caption system audio or other apps (beyond the current tab) - Send system audio into Zoom or Teams - Record desktop audio in OBS - Mix mic + music into one output - Capture app-specific sound Below are reliable tools and how to use them. Windows On Windows, the two most common tools are: - **VB-Audio Virtual Cable (VB-CABLE)** — Free, simple virtual cable. Creates a virtual input/output pair so you can route audio between applications. - **Virtual Audio Cable (VAC)** — Commercial option with more routing flexibility. After installing, select the virtual output as your system playback device or route specific apps to it, then choose the virtual input as the microphone source in Live Caption (Microphone mode). macOS On macOS, the most powerful and flexible solution is: - **Loopback** (by Rogue Amoeba) — Combines audio from multiple apps and devices into one virtual source. Lets you mix system audio, mic, and specific apps into a single input. --** Blackhole** After setting up Loopback, choose the Loopback device as your microphone in Live Caption (Microphone mode) to caption the routed audio.