Capture
Screenshots, source windows, Steam Deck overlays, browser text, subtitles, articles, lyrics, and scripts.
2025-present / Acorn Talk
Acorn Talk
A year-long attempt to make native Japanese content feel reachable: capture what you want to understand, save the words that matter, and review them in the context where they became real.
The Beginning
The problem was not vocabulary. It was motivation.
I was studying Japanese with the usual beginner tools, but the material I actually cared about was somewhere else: games, anime, manga, YouTube, songs, and conversations. The generic vocabulary path felt disconnected from the reason I wanted to learn.
The turning point was a small measurement. The first episode of Spy x Family had roughly 600 unique words. Out of about 2,000 words I knew from textbooks and WaniKani-style study, only about 100 showed up in that episode. Even the next thousand generic words would only add about 80 more. I did not need a more heroic grind. I needed a study plan aimed at the content I wanted next.
The Loop
Native content becomes study material
Understand
OCR, Japanese tokenization, deconjugation, dictionary matching, LLM-assisted context, translation, and ambiguity recovery.
Save
Selected words keep their sentence, screenshot, reading, sense, audio state, source pack, and occurrence metadata.
Review
Tango materializes notes, items, contexts, and cards so learners build quick kanji and word recognition.
Tango
The memory layer
Tango is organized around packs. A pack might be a game, a manga volume, an anime episode, a textbook chapter, an article, a song, or a pile of Lens captures. The pack keeps motivation local: study this because it helps you understand that.
The review system is opinionated. Kanji cards train quick recognition. Word cards train readings and contextual understanding. Slow correct answers still matter because the goal is automaticity: seeing a word and recognizing it before the friction makes the original content feel like work.
Lens
The capture layer
Lens began as the missing piece for Japanese games on Steam Deck. It needed to let me stay in the game, inspect a line, hear pronunciation, save a word, and continue playing. Later, Desktop Lens brought the same idea to any window on Windows, macOS, and Linux.
The important shift was from automatic extraction to explicit saving. Lens can analyze a lot of text, but only the words you choose become study material. That keeps the user's pack from turning into an undifferentiated backlog.
Product Surface
A suite, not a single app
Create
Paste or upload Japanese text, inspect detected words line by line, then save the pieces worth reviewing.
Review
Study sessions keep readings, context sentences, kanji breakdowns, and audio close to the answer.
Audio
Native speaker recordings are preferred, synthetic fallback is labeled, and bad generated audio can be detected and replaced.
Engineering
Making lookup feel instant changed the architecture
The original Lens pipeline was designed for batch screenshots: OCR, LLM classification, token refinement, JMDict matching, disambiguation, extraction writing, pack inclusion, and materialization. It was powerful, but it was too slow for a desktop experience where a dictionary popup has to appear as the pointer moves.
The desktop direction moved the model out of the hot path. OCR runs once, character boxes and offsets support hit testing, a local dictionary artifact handles fast lookup, and the server only materializes the words the user saves. The hard part was not calling a faster model. It was deciding where the model should not be.
Screenshot or text
Window capture, Steam Deck overlay, subtitles, lyrics, scripts, articles, or manual text.
OCR and Japanese NLP
Vision OCR, bounding boxes, Sudachi, dictionary forms, deconjugation, and ambiguity handling.
Words in context
Surface forms, selected senses, readings, sentence context, screenshots, audio, and source packs.
SRS materialization
Notes, items, contexts, kanji cards, word cards, review history, and pack-specific progress.
What Shipped
One year of compounding pieces
The motivation problem
Generic study lists were not getting me closer to the specific Japanese I wanted to read, watch, and play.
Tango takes shape
Packs became the core abstraction: a way to keep vocabulary, kanji, sentences, audio, and progress tied to one real source.
Open beta and Lens for Steam Deck
Tango opened to everyone, and Lens gave Japanese games a read-and-save overlay on Steam Deck.
Lens for Desktop
Lens expanded to Windows, macOS, and Linux with hotkey capture, OCR, dictionary lookup, audio, and saving into Tango.
One year in
The project had grown from personal study tooling into a product suite, a corpus, a desktop app, a Steam Deck plugin, and a learning workflow I kept using.
Why It Worked
It made Japanese part of the day I wanted to have
The best sign was not a metric dashboard. It was wanting to come back. Instead of studying generic words for a future payoff, I could study the words that made a game scene, a song lyric, a manga page, or a news article more comprehensible right now.
By June 2026, other users were starting to show the same pattern. One Steam Deck user wrote after finishing NieR: Automata in Japanese that they expected this kind of resource not to exist, installed Lens, and found it useful enough to call it a game changer. That is the project in miniature: lower the lookup friction, preserve the context, and help the learner keep going.
Gallery / Memories
Pieces of the Acorn Talk system
A few pieces of the product surface: desktop lookup, Steam Deck lookup, Tango study, kanji feedback, audio quality work, and the gallery where captures become packs.
Media preview
Lens desktop gallery
LensCaptures are organized into packs so a game, manga volume, show, or article can become a study project.
Gallery item 1 of 8