NanumPlayer

Watch anything in your own language. NanumPlayer listens to the dialogue, translates it on the fly and shows AI subtitles while you watch.

Download Get started GitHub

NanumPlayer playing an English test clip. Above the bottom playback bar, an AI Spanish translation, 'He mantenido esta lámpara encendida todas las noches durante 30 años', appears over the original English line, 'I have kept this lamp burning every night for 30 years.' The status chips show speech recognition on the GPU and the translation model. — Example: English dialogue in a test clip, shown with an AI Spanish translation above the original line.

Download

Download for WindowsPortable ZIP · SHA-256 checksums · All releases

Windows 10/11 (64-bit), CPU with AVX2. Free and open source (GPL-2.0-or-later).

Every release lists the SHA-256 of each file. Compare it with Get-FileHash before you run the installer.

Verify in PowerShell: Get-FileHash .\NanumPlayer-Setup-<version>.exe -Algorithm SHA256

The installer does not include speech models or the optional Blu-ray/DVD decryption components. See Getting started for what to download next.

New here? Getting started takes you from the installer to your first AI subtitles. Everything runs on your PC; the optional parts (speech models, a translation model) are downloaded only when you choose them.

What it does

Real-time AI subtitles

Speech recognition (Whisper) turns the audio into text, a translation model of your choice turns it into your language, and the result appears as a normal subtitle track, ahead of playback.

Files, discs and the web

Local video files, Blu-ray and DVD, YouTube and videos on ordinary web pages. Subtitles work the same everywhere.

Subtitle files without playing

Generate SRT files for a whole video from speech, from a text subtitle track or from disc image subtitles (OCR), with one translated file per language.

Glossary and style

Character and place names stay consistent. Review and edit the glossary, choose polite, casual or natural speech, and existing subtitles are re-translated.

Sharper pictures

GPU upscaling (high-quality scaling, NVIDIA RTX Video Super Resolution, your own shaders) and built-in picture-quality presets for anime and compressed video.

Private by default

No playback history, logs or subtitle caches are kept unless you ask for them. Web and YouTube windows always run in InPrivate mode. No telemetry.

See it

NanumPlayer's playback bar with status chips for speech recognition on the GPU, the large-v3-turbo model and the local translation model, over a lighthouse scene. — The playback bar shows how the selected AI subtitle track is made: speech model, GPU or CPU, and the translation model.

The Subtitles list with entries for AI translation, translation plus original, the original-language recognition track, Off and Load a subtitle file. — One subtitle list for AI tracks, disc and file subtitles.

The NanumPlayer settings menu opened with Ctrl+N, listing interface language, AI component manager, subtitle, upscaling, picture quality and translation style. — Ctrl+N opens every setting, with search.

The NanumPlayer start screen with buttons for Open file, Open Blu-ray, YouTube, Play from a web page and AI subtitle settings. — The start screen. Interface languages: English, Korean, Japanese, Simplified and Traditional Chinese, and Spanish.

Good to know

Windows only. Windows 10 or 11, 64-bit, with a CPU that supports AVX2. An NVIDIA GPU is optional and speeds up speech recognition.
You need a translation model. Subtitles are translated by an Ollama-compatible server, local or cloud. See Getting started.
Disc decryption is a separate, optional download. Unprotected discs, folders and ISO images play with the normal installer. See Optional Blu-ray/DVD decryption components.
Open source. NanumPlayer is a modified mpv (GPL-2.0-or-later) with whisper.cpp (MIT). See License and notices.

FAQSmartScreen, GPU requirements, offline use and languages.

PrivacyWhat stays on your PC and what connects to the network.