Methodology and Sources

Last verified April 2026

A linguistic reference is only as trustworthy as its sourcing. This page explains where the content comes from, the standards used, and the specific licences and attributions for the data on this site.

Sources

Every linguistic claim on this site is verifiable against at least one of:

Where claims conflict between sources, both readings are noted, with one marked as standard, formal, or rare. The site does not rely on AI-generated content, anecdotal Reddit threads, or anime as a source for register or idiom (anime uses fictional speech patterns).

Romanisation standard

Modified Hepburn. Macrons used for long vowels (jū, kyū, chō). Long vowels are not doubled (jū not juu). Apostrophes used to disambiguate where needed (kin’yōbi, not kinyōbi). The ん (n) sound is written n (not m) before b, m, p (sanpo, not sampo).

Audio policy

The audio buttons use the Web Speech API with the ja-JP voice. Quality varies by browser and device. Where the API is unavailable, the page indicates "audio unsupported". The site does not voice-clone, scrape Forvo, or synthesise audio server-side. A future commission of native-speaker MP3s will supplement the most-used entries.

Image policy

No DALL-E or generative imagery for kanji. Generative AI is unreliable with kanji shapes. Every kanji on this site is rendered as Unicode text in Noto Sans JP and Noto Serif JP, two open-source fonts maintained by Google. Stroke counts are referenced from KanjiVG (CC-BY-SA 3.0); future stroke-order animations will use KanjiVG SVG data with full attribution.

Last-verified policy

Every page on this site carries a "Last verified April 2026" stamp under the H1. The stamp updates when the page content has been re-checked against the source list above. It tells search engines and human readers when the linguistic content was last reviewed.

KanjiVG attribution

Stroke-count data and any future stroke-order diagrams on this site are derived from KanjiVG, a project by Ulrich Apel. KanjiVG is released under the Creative Commons Attribution-Share Alike 3.0 licence (CC-BY-SA 3.0). We thank the contributors and credit the project here, on the relevant content pages, and in the project log.

Disclaimer

This site is an independent language-learning reference. It is not affiliated with Tofugu, WaniKani, Busuu, Rosetta Stone, NHK, the Japan Foundation, or any Japanese-language educational publisher. Romanisation follows modified Hepburn standard. Audio uses the Web Speech API (ja-JP voice) where browsers support it; pre-recorded native audio supplements key entries.

Corrections

Spotted an error? Email corrections to corrections@numbersinjapanese.com. Linguistic claims should be reproducible from one of the sources above; if you find a discrepancy, please reference the source you believe is correct.

Footer credit

Built by Digital Signet. Last verified April 2026.