Loading…
CompareOpen in Google Play
This app isn't in any of our charts yet.
Create a family language folio, record and review speech, compare parallel forms, choose entry-level contributions, and export original audio with open lexicon formats.
Record family speech and shape it into an open, growing dictionary.
Liafuan helps two relatives begin a dictionary for a language their family speaks but may not write in a standard way. One person speaks naturally. The other adds a rough meaning, an optional spelling, and a private note. The phone studies sound and silence on the device. It suggests sentence boundaries and rough timing for the words you enter. Every suggestion stays editable, and the app never marks one family form as the correct form. A growing dictionary page shows what the family has recorded and which conversation areas still have space. The next prompt responds to that coverage, recent recordings, skipped prompts, acoustic variety, and unresolved forms. Prompts are conversation cues, not lessons or tests. Different forms remain visible as parallel branches. Several recordings from one household improve that household’s evidence, but count as one household witness. This keeps repeated recordings from looking like independent agreement. Recording, analysis, review, search, playback, editing, local variant grouping, and export continue without the contribution service. App-specific fingerprint/face and PIN locks are available independently. You can choose to contribute one entry at a time to a shared catalog. Audio is never included unless you select it separately for that entry. A shared entry can be withdrawn later. Shared forms include reporting and household-blocking controls, and the app provides child-safety and shared-data deletion routes. One export writes the original M4A recordings together with structured JSON, UTF-8 CSV, ELAN-compatible EAF annotations, a readable manifest, and SHA-256 checksums. The archive can be opened without Liafuan. Liafuan is for building a family reference from living speech. It does not teach lessons, translate languages, score speakers, create public profiles, or declare a single authoritative form.