DualPiP – PiP Player | ASR Live Captions · Subtitles & AI Translation

PiP player with ASR live captions, bilingual subtitles, AI translation, and danmaku. Supports YouTube, Netflix, Disney+, and more.

What can the DualPiP Picture-in-Picture player do? Play any video in a floating window with full controls and these features: Subtitles: • ASR Live Captions (PRO) — Auto-generate AI captions via speech recognition, powered by Deepgram or local Whisper • Bilingual subtitles — Display two languages simultaneously with multiple translation engines • Online search — Find and download from OpenSubtitles, SubDL, and more • Local import — Load SRT/VTT/ASS/SSA/SUB files • Style customization — Adjust font, size, color, position, and background Translation: • AI engines — 30+ providers including OpenAI, Claude, Gemini, and DeepSeek for context-aware results • Traditional engines — Google Translate, Microsoft Translator, and Yandex • Batch pre-translation — Process all content ahead of playback for zero wait Playback: • Full controls — Play/pause, volume, seek, speed, and frame-by-frame • Gestures — Double-tap to seek, swipe for progress/volume/brightness, long-press for speed boost • Timeline thumbnail preview and customizable keyboard shortcuts Learning: • Learning Mode — Subtitle list panel with split and collapsed layouts, double-click to jump • AB Loop — Multi-loop management with audio fade-in/out and precise timing • Smart merge — Automatically combine fragments into complete sentences Interactive: • Danmaku — Live comments from Bilibili, NicoNico, YouTube Live, Twitch, and Kick • Side panel for chat messages and one-click video screenshots How does ASR speech recognition work? ASR is a PRO feature that auto-generates real-time AI captions for any web video. It supports Deepgram cloud (22+ languages, <300ms latency) and local Whisper servers. Two capture modes: • Streaming (WebSocket) — Ideal for livestreams, video calls, and real-time content • Pre-download batch (HTTP) — Generate complete captions before playback for published videos Uses a BYOK (Bring Your Own Key) architecture — requests go directly from your browser to the provider with no middleman. Generated captions fully integrate with Learning Mode, AB Loop, and AI translation for bilingual output. What sites are supported? Works with any website that has an HTML5 video element, including YouTube, Netflix, Disney+, and Bilibili for streaming; Coursera and Udemy for education; Twitch for live streams; and Zoom web for meetings. Who is it for? • Language learners — Real-time foreign-language captions with AI translation and AB loop practice • Students — Auto-generated captions for online courses with learning tools • Viewers of unsubtitled content — Captions for livestreams, uploads, and niche videos • Anime, drama, and live stream fans — Danmaku overlay, bilingual display, speed control • Multitaskers — Always-on-top floating window for working while watching Install DualPiP for a more efficient video watching and language learning experience.