60+ languages, a dozen families.
Laesi's language list is the whole point. We prioritize coverage in the families mass-market reader apps ignore, and cover the mainstream ones too. Click any language for reading samples, a suggested library, and exactly what kind of lemmatization it has today.
Nordic & North Germanic
The living and medieval languages of Scandinavia, Iceland, and the Faroes.Icelandic preserves Old Norse grammar more faithfully than any other living language. Laesi supports it today with Wiktionary lookups; BÍN-grade lemmatization is on the roadmap.
Spoken by 70,000 people on 18 islands. No major reader app supports it. Laesi does, with a GiellaLT morphological analyzer and Wiktionary lookups.
Bokmål is the written standard roughly nine Norwegian writers in ten reach for. Full spaCy lemmatization on day one.
Built from Norway's rural dialects by Ivar Aasen in the 1850s. If you read Tarjei Vesaas or Jon Fosse, you want Nynorsk.
Danish grammar is simple; Danish pronunciation is famously not. Laesi focuses on what you study in text: vocabulary, idioms, phrasal verbs.
Swedish has the largest literary output of the Nordic languages. Laesi handles its compound nouns, particle verbs, and two genders through spaCy.
The medieval language of Iceland, Norway, and the Viking diaspora. Almost no reading app supports it. Laesi gives it a GiellaLT analyzer. CLTK, a better fit for Old Norse, is on the roadmap.
A conservative North Germanic variety of Älvdalen, Sweden, that kept features lost everywhere else. Surface-form lookup today.
West Germanic
German and its relatives, from Standard German to the regional and minority varieties.Full spaCy lemmatization across four cases, three genders, and the strong verbs. The base model also powers Swiss German and Bavarian.
Full spaCy lemmatization. Handles Dutch compounds, separable verbs, and diminutives.
Lemmas are stored as Standard German forms, there's no standard Swiss German orthography, so lemmatization is partial but useful.
Lemmas are stored as Standard German forms. Partial lemmatization, full reading and lookup support.
German Low German (SASS orthography). Surface-form lookup against Wiktionary's Low German section.
Low German as written in the Netherlands (NSS). A separate language entry from German Low German.
Surface-form lookup today. A national language with a small but growing written corpus.
The closest living relative of English. Surface-form lookup against Wiktionary.
A cluster of endangered Frisian dialects on the German North Sea coast. Surface-form lookup.
The Germanic sister language of English, spoken in Lowland Scotland and Ulster. Surface-form lookup.
Romance
The daughters of Latin, with full lemmatization across the major standards.Full spaCy lemmatization with elision, contraction, and verb-conjugation handling.
Full spaCy lemmatization. Handles enclitic pronouns and the full verb paradigm.
Full spaCy lemmatization with articulated prepositions and clitic handling.
Full spaCy lemmatization covering both European and Brazilian orthography.
Full spaCy lemmatization. EPUB import normalizes legacy cedilla forms to comma-below diacritics.
Full spaCy lemmatization for the language of Catalonia, Valencia, and the Balearics.
Surface-form lookup today for the Romance language of northwestern Spain.
Slavic
East, West, and South Slavic, with full lemmatization for the major standards and script-aware handling.Full spaCy lemmatization across Polish's seven cases and complex consonant alternations.
Full spaCy lemmatization with case, aspect, and Cyrillic handling.
Full spaCy lemmatization for Ukrainian's seven cases and verbal aspect.
Full spaCy lemmatization. Handles Bulgarian's postposed definite article.
Full spaCy lemmatization, including the rare dual number.
Full spaCy lemmatization for the South Slavic language of North Macedonia.
Full spaCy lemmatization. The Croatian model also backs Serbian (Latin).
Lemmatized via the Croatian model with a Cyrillic-leak guard. A dedicated Serbian model is planned.
A separate language entry from Serbian Latin. Surface-form lookup; CLASSLA lemmatization planned.
Surface-form lookup today. A Stanza-based lemmatizer is planned.
Surface-form lookup today. A Stanza-based lemmatizer is planned.
Surface-form lookup today. A Stanza-based lemmatizer is planned.
Surface-form lookup. A separate entry from Bosnian Cyrillic; CLASSLA lemmatization planned.
Surface-form lookup. A separate entry from Bosnian Latin.
Baltic
The conservative Indo-European languages of the eastern Baltic.Uralic & Finno-Ugric
Finnic, Sámi, and their relatives, agglutinative, morphology-heavy, and underserved.Finnish has a reputation for being hard because its words shape-shift. Voikko ties every form you meet back to one lemma, so you can tell which of the fifteen you actually recognise.
North Sámi is spoken across the Arctic reaches of Norway, Sweden, and Finland. Laesi supports it with GiellaLT, the analyzer the Sámi language community built.
Full GiellaLT morphology. Wiktionary coverage is thin but the analyzer is solid.
Full GiellaLT morphology for the southernmost Sámi language.
Full GiellaLT morphology for the Sámi language of the Inari region in Finland.
Full GiellaLT morphology for one of the most endangered Sámi languages.
A Finnic language of northern Norway. Full GiellaLT morphology, thin Wiktionary.
A Finnic minority language of the Torne Valley in Sweden. Full GiellaLT morphology.
Surface-form lookup today; a Stanza lemmatizer is planned. Agglutinative, so inflected hit rates are low until then.
Surface-form lookup today; a Stanza lemmatizer is planned. Highly agglutinative.
Celtic
Goidelic and Brythonic languages with initial mutations and VSO grammar.Irish's grammatical mutations (séimhiú, urú) break naive word lookup. Laesi routes Irish through a GiellaLT analyzer that normalizes them back to dictionary form.
Laesi supports Welsh today with Wiktionary surface-form lookup. A morphological analyzer is being researched, Welsh has no GiellaLT model, so we're evaluating the alternatives.
Hellenic & Mediterranean
Greek, Maltese, and Albanian, the Mediterranean languages other readers skip.Full spaCy lemmatization for Modern Greek, including the polytonic-to-monotonic normalization.
The only Semitic language written in Latin script and an EU official language. Full spaCy lemmatization.
Surface-form lookup today for the sole surviving branch of its own Indo-European family.
Turkic
Agglutinative Turkic languages, surface-form lookup today, lemmatization on the roadmap.Surface-form lookup today. Highly agglutinative, so a dedicated lemmatizer is a priority for inflected hit rates.
Surface-form lookup today for the Turkic language of Azerbaijan and northwestern Iran.
Indigenous & Endangered
Languages with small speaker communities and active revitalization: exactly the ones the big apps will never release.Kalaallisut builds whole sentences into single words. Laesi supports it with a GiellaLT analyzer, root-level by default, morpheme-level optional.
Southwestern Ojibwe (ISO ciw): the Minnesota dialect cluster, and the variety with a GiellaLT analyzer. Built from source; Wiktionary coverage is very thin.
Central Ojibwa (ISO ojc): the Ontario dialect cluster, and the variety with a GiellaLT analyzer. Built from source; Wiktionary coverage is very thin.
World languages
Major and minor languages from beyond Europe, readable today with surface-form lookup.Surface-form lookup today. Light morphology means hit rates are already good.
Surface-form lookup today for the language of Malaysia, Brunei, and Singapore.
Surface-form lookup today for the national language of the Philippines.
Surface-form lookup today for East Africa's lingua franca.
Surface-form lookup today for the French-based creole of Haiti.
Surface-form lookup today. Analytic grammar means inflection is rarely an obstacle.
Surface-form lookup for the world's most successful constructed language.
Classical & Historical
Dead languages with living literatures. Coming as the CLTK analyzers land.Latin learners juggle Whitaker's Words, Alpheios, Logeion, and Anki. Laesi will bring reading, lookup, SRS export, and progress tracking into one place.
The language of Alfred, of Beowulf, of the Anglo-Saxon Chronicle. CLTK integration will handle Old English's strong and weak declensions.
Middle English spelling is famously inconsistent. Laesi's surface-form handling will tie spelling variants to one another so you study vocabulary, not orthographic accidents.
Don't see your language?
Add it yourself: Laesi's custom-language tool handles minority dialects, conlangs, and niche historical varieties. Or email us: if there's a GiellaLT, CLTK, spaCy, or Stanza analyzer we can wire in, we'll prioritize it.