Hi Zanga, A form analysing against a monolingual transducer but not against its version in a translation pair is almost always due to dictionary trimming: https://wiki.apertium.org/wiki/Automatically_trimming_a_monodix https://wiki.apertium.org/wiki/Why_we_trim
This in turn is often due to a missing or incorrect entry in the bilingual dictionary, since the monolingual analysers are trimmed to include only stems in the bilingual dictionary. In your case, I see you consider this a class 12 noun in the monolingual analyser. In the bilingual dictionary you have the same lemma and POS with a subcategory tag of <cl13>. That'll definitely lead to this kind of output. If that doesn't fix the problem, double check that the yao generator has the exact stem as in in the bilingual dictionary. Also feel free to hop on IRC—we can support you in real time there! https://wiki.apertium.org/wiki/IRC -- Jonathan On Sun, Aug 23, 2026, 02:12 Zanga Chimombo <[email protected]> wrote: > I am getting the following (expected and correct) result in > apertium-nya monolingual directory: > echo kamtengo kakakulu | lt-proc -w nya.automorf.bin > ^kamtengo/mtengo<n><cl12><sg>$ ^kakakulu/kulu<adj><sg><cl12>$ > > However, in the apertium-nya-yao bilingual directory, it seems to have > become unknown and is marked with an *: > echo kamtengo kakakulu | lt-proc -w nya-yao.automorf.bin > ^kamtengo/*kamtengo$ ^kakakulu/kulu<adj><sg><cl12>$ > > Where should I be looking to debug this please? > > Zanga. > > > _______________________________________________ > Apertium-stuff mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/apertium-stuff >
_______________________________________________ Apertium-stuff mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/apertium-stuff
