Batch-mines Japanese vocabulary from anime and YouTube into Anki cards. Given a season folder or a YouTube URL, it produces cards containing screenshots, sentence audio, furigana, pitch accent, and frequency data.
Suited to batch processing after viewing, rather than real-time lookup during playback (the asbplayer and Yomitan workflow).
 |
 |
 |
Generated from video and subtitle files. Each card contains a screenshot, sentence audio, furigana, and definition.
- Parse subtitles: tokenize Japanese text with MeCab morphological analysis.
- Filter words: keep content words (nouns, verbs, adjectives, adverbs); drop words already in your Anki collection or on your blacklist.
- Extract media: capture screenshots and audio clips from the video at each subtitle's timestamp via ffmpeg.
- Fetch definitions: look up English definitions from JMdict (offline) or the Jisho API.
- Create cards: batch upload to Anki via AnkiConnect.
- Lapis-compatible cards with furigana, pitch accent, and word frequency fields.
- YouTube support: paste a URL, mine the video.
- Queue a folder of episode/subtitle pairs for sequential processing.
- Offline JMdict dictionary with Jisho API fallback.
- Preview and curate the word list before any cards are created.
- Parallel ffmpeg extraction for screenshots and sentence audio.
- Analytics dashboard with history, undo, and series difficulty rankings.
- Four themes (Light, Dark, Sakura, Tokyo Night) plus custom JSON themes.
- ffmpeg on PATH.
- macOS:
brew install ffmpeg - Ubuntu/Debian:
sudo apt install ffmpeg - Windows: download from ffmpeg.org and add to PATH.
- macOS:
- Anki with the AnkiConnect add-on. In Anki: Tools → Add-ons → Get Add-ons, paste code
2055492159, restart.
Grab the installer for your platform from the latest release:
| Platform | Installer | Portable |
|---|---|---|
| Windows | AnkiMiner-*-Setup.exe |
AnkiMiner-Windows-x86_64.zip |
| Linux (Debian/Ubuntu) | anki-miner_*_amd64.deb |
AnkiMiner-*-Linux-x86_64.AppImage |
| Linux (other) | — | AnkiMiner-Linux-x86_64.tar.gz |
| macOS (Apple Silicon) | — | AnkiMiner-macOS-arm64.tar.gz |
No Python required. Installers and portable archives bundle all dependencies.
Install from PyPI (Python 3.10+)
pipx install anki-miner # or: pip install anki-minerInstall from source
git clone https://github.com/0xzerolight/anki_miner.git
cd anki_miner
pip install .After installing, launch Anki Miner from your Start Menu, Applications folder, or app menu. If you installed from PyPI or source, run anki_miner_gui from a terminal. A desktop shortcut is created on first launch; re-run it from Tools -> Create Desktop Shortcut... inside the app.
Anki must be running with AnkiConnect installed before mining starts.
Tabs:
- Single Episode: mine one video/subtitle pair with file selectors and progress tracking.
- Batch Processing: queue multiple series for sequential processing.
- YouTube: paste a URL, fetch metadata, then mine.
- Analytics: history, series difficulty, milestones.
- Settings: Anki connection, media extraction, dictionary, word filtering. Saved to
~/.anki_miner/gui_config.json.
Anki Miner uses the Lapis note type fields by default. For custom note types, rename the fields in Settings/Anki.
- Download the latest
.apkgfrom Lapis releases. - In Anki: File → Import and select the
.apkg.
Default field mapping:
| Anki Miner Field | Note Field | Content |
|---|---|---|
| word | Expression | Dictionary form of the word |
| sentence | Sentence | Original subtitle line |
| definition | MainDefinition | English definitions |
| picture | Picture | Screenshot from the video |
| audio | SentenceAudio | Audio clip of the sentence |
| expression_furigana | ExpressionFurigana | Word with furigana reading |
| sentence_furigana | SentenceFurigana | Sentence with furigana reading |
| pitch_position | (unmapped) | Pitch accent position number |
| pitch_category | (unmapped) | Pitch accent category |
| frequency | (unmapped) | Word frequency rank |
Fields marked (unmapped) have no default Lapis mapping. Map them in Settings if your note type has equivalents. Any note type with the required fields works.
For fast offline lookups:
mkdir -p ~/.anki_miner
wget -O ~/.anki_miner/JMdict_e.gz http://ftp.edrdg.org/pub/Nihongo/JMdict_e.gz
gunzip ~/.anki_miner/JMdict_e.gzWithout JMdict, lookups fall back to the Jisho API (slower, online, rate-limited).
Paste a URL, click Fetch Info to probe metadata (title, duration, subtitle availability), then click Mine. The fetch downloads the video and its Japanese subtitle track into a per-run temporary directory, then passes both files to the same pipeline used for file-based mining.
Auto-captions are accepted only when native Japanese. Tracks that YouTube generates by machine-translating from English are rejected, since mining them yields unusable results. Native auto-captions remain lower quality than manual subtitles because they lack sentence boundaries.
Gotchas:
- Bot-detection prompts: if YouTube asks "Sign in to confirm you're not a bot", open Settings -> Cookies -> Browser and pick Firefox or Chrome. yt-dlp pulls cookies from that browser's profile on every fetch.
- Age-restricted videos: same fix.
- Max duration: defaults to 120 minutes. The probe aborts before downloading if the video is longer. Adjust in Settings.
Anki Miner checks GitHub for new releases on startup (toggle in Settings). When an update is available, a banner offers a one-click download of the asset that matches your install: .deb for Debian/Ubuntu, .AppImage for AppImage, the Inno installer on Windows, the macOS arm64 archive, or the release page for pip/source installs. "Skip this version" suppresses the prompt for that release; the next release prompts again.
| Issue | Solution |
|---|---|
| "Cannot connect to Anki" | Start Anki and ensure AnkiConnect is installed. |
| "Deck not found" | Create the deck in Anki or update the deck name in Settings. |
| "Note type not found" | Import Lapis (see above) or configure your own in Settings. |
| "ffmpeg not found" | Install ffmpeg and add it to PATH. |
| "JMdict file not found" | Download to ~/.anki_miner/ (see above) or disable offline dictionary. |
| Audio is wrong language | The tool tries Japanese audio tracks first, then falls back to the default. |
| Subtitles out of sync | Use the subtitle offset control in the GUI. |
Report bugs in Issues.
See CONTRIBUTING.md for development setup.
New feature ideas in Discussions.
GNU General Public License v3.0. See LICENSE.