language-immersion-tv

Language Immersion TV

Overview

Language Immersion TV transforms the media you already watch into a personalized language-learning curriculum. It parses subtitle files (.srt, .vtt) from movies and TV shows, extracts the most useful vocabulary, builds frequency-ranked word lists, identifies idioms and multi-word expressions, and exports ready-to-study decks in Anki and CSV format.

The core insight: you learn faster from content you actually enjoy. Watching a show you love in your target language makes vocabulary stick because you have emotional context, visual cues, and narrative motivation.

When to Use

• You're learning a language and want to learn from shows you actually watch
• You have subtitle files (.srt/.vtt) and want to extract useful vocabulary
• You want frequency-based vocabulary decks built from real dialogue
• You want to build a "watch and study" curriculum from a TV series
• Don't use for: learning a language from zero — use a structured beginner course first, then use this to accelerate from A2 onward

How It Works

1. Parse subtitles — Read .srt/.vtt files, clean timestamps, extract clean dialogue text
2. Tokenize & lemmatize — Split into words, normalize to dictionary forms
3. Frequency analysis — Rank words by occurrence; identify high-value words you'll encounter again
4. Phrase extraction — Detect common collocations and idioms (e.g., "of course", "what's up", "never mind")
5. Difficulty scoring — Classify words as A1/A2/B1/B2/C1 based on frequency
6. Deck export — Generate Anki-compatible TSV, CSV, or JSON flashcards with context sentences

Quick Start

# Analyze a single subtitle file
python scripts/immersion.py analyze subs/movie.srt --language en

# Build a vocabulary deck from a whole season
python scripts/immersion.py build-season subs/ --language es --output deck.json

# Export to Anki-importable TSV
python scripts/immersion.py export subs/movie.srt --language fr --format anki --output cards.tsv

# Compare vocabulary across multiple episodes to find common words
python scripts/immersion.py compare subs/ --language de --top 100

Supported Languages

Built-in stopword lists and basic lemmatization for: English, Spanish, French, German, Italian, Portuguese.

Other languages work but without smart filtering — all words pass through.

Workflow: Building a Study Curriculum from a TV Series

Step 1: Collect subtitle files

Gather .srt or .vtt files for the series you're watching in your target language.

Step 2: Build a master frequency list

python scripts/immersion.py build-season ~/Downloads/breaking-bad-subs/ --language en --output bb_master.json

This identifies the most frequent words across the whole series — the vocabulary you'll encounter repeatedly.

Step 3: Export episode-by-episode decks

for f in ~/Downloads/breaking-bad-subs/*.srt; do
  python scripts/immersion.py export "$f" --language en --format anki --output "${f%.srt}.ts"
done

Step 4: Study before watching

Before watching an episode, review its deck (50–100 cards). Then watch the episode and notice the words in context.

Step 5: Compare and refine

python scripts/immersion.py compare ~/Downloads/breaking-bad-subs/ --language en --top 200

Words that appear in many episodes are your highest-value targets.

CEFR Level Estimation

The tool estimates word difficulty based on frequency percentile:

Common Pitfalls

1. Studying every word. Focus on words that appear 3+ times across a series. One-off words are low ROI.
2. Ignoring context. Always study words with their original subtitle sentence — context is what makes them stick.
3. Starting with hard content. If you're A2, don't start with medical dramas. Begin with sitcoms and cartoons where speech is slower and vocabulary is everyday.
4. Forgetting to set the language. The tool needs --language for proper stopword removal. Without it, "the", "a", "is" will dominate your deck.
5. Using machine-translated subtitles. Always use native-language subtitles, not auto-translated ones. The vocabulary is authentic.

Verification Checklist

• immersion.py analyze subs/movie.srt --language en prints word frequency stats
• immersion.py export subs/movie.srt --language en --format csv produces a CSV file
• immersion.py build-season subs/ --language es produces a combined frequency list
• immersion.py compare subs/ --top 50 shows cross-episode common words
• Anki TSV output imports cleanly into Anki (tab-separated: word, context, translation-placeholder)

References

• references/methodology.md — the linguistics and cognitive science of learning from comprehensible input
• references/cefr-levels.md — CEFR framework explained, with media recommendations per level