Dear colleagues

Thanks a lot to Kilian Evang, Gilles Sérasset and Adriana Stan (who gave me the link https://github.com/open-dict-data/ipa-dict) and Camila Buitrago for your kind help.

I also found https://pypi.org/project/ipapy/

As I said, I want to find neologisms for Andean Amazonian languages, for sure the first source of neologisms are the same languages but is unavoidable (and maybe positive!) try neologisms coming from other languages. Against the "common sense" of getting neologisms only from spanish, I am looking for neologisms from any language under the condition that the neologisms "sound like" the words of the target languages. Since natural languages rarely have perfectly phonemic orthographies, scripting is not useful for my research, trying speech recognition is too much, so I bet on checking out IPA representations of as many foreign languages' words as possible.

Thanks to your links, now I have more than 500k words written in IPA format. Next, I'll define and encode the "similar-sounding" rules, and then iterate through all of those words to find the eligible ones. Probably, you already notice my goal is not to propose a few but a massive amount of neologisms. As far as I know, there is no background, but if you know some, I'd appreciate your input. I expect to unleash the code and a paper draft this month.

Best regards

Luis

El jue, 15 sept 2022 a las 9:17, Kilian Evang (kilian.evang@gmail.com) escribió:

Hi Luis,

Another resource you might want to look into is WikiPron:

https://github.com/kylebgorman/wikipron

Cheers, Kilian

Am Do., 15. Sept. 2022 um 15:58 Uhr schrieb Gilles Sérasset < Gilles.Serasset@univ-grenoble-alpes.fr>:

...

Hi Luis,

Don’t know if this could be useful to you, but currently, the DBnary dataset contains phonetic (IPA) transcription of many entries.

DBnary is linked data and can be explored through its public endpoint using SPARQL language: http://kaiko.getalp.org/sparql

For instance the following query will tell you how many phonetic reps are available in which languages.

select ?lang count(?pr) where { [] ontolex:phoneticRep ?pr. BIND (lang(?pr) as ?lang) } GROUP BY ?lang ORDER BY DESC(COUNT(?pr))

This will give you a long table (I only include the first lines (results are order on the number of phoneticRep).

langcallret-1

fr-fonipa

2657875

en-fonipa

663697

ru-fonipa

389891

de-fonipa

230875

fi-fonipa

199269

es-fonipa

187090

la-fonipa

171134

it-fonipa

154881

pl-fonipa

136446

sh-fonipa

116478

pt-fonipa

90199

ca-fonipa

86385

eo-fonipa

84626

avk-fonipa

73459

es-ipa

72652

vi-fonipa

72147

As the data is continuously extracted from wiktionaries, the numbers will evolve (and as several language extractors do not yet extract the phonetic representation, feel free to file a feature request on DBnary bug tracker).

More info at :

http://kaiko.getalp.org/about-dbnary/

Regards,

Gilles,

On 7 Sep 2022, at 16:26, Luis Camacho Caballero camacho.l@pucp.edu.pe wrote:

Dear colleagues

I'm devoted to the revitalization and massification of the Andean Amazonian native language with computational processing as a key enabler.

Among the many tasks to do, nowadays I'm dealing with the creation of neologisms. That is why I'm looking for the larger multilingual dictionary of phonetic spelling, even better if that database includes asian languages (mandarin, japanese, korean, hindi, urdu, etc).

If you have this kind of database, I kindly ask you for bring me access, if you don't, I'd appreciate any clue about where and/or how access to it

Kind regards

Luis Camacho https://orcid.org/0000-0001-6569-550X

---

---

Corpora mailing list -- corpora@list.elra.info https://list.elra.info/mailman3/postorius/lists/corpora.list.elra.info/ To unsubscribe send an email to corpora-leave@list.elra.info

---

Corpora mailing list -- corpora@list.elra.info https://list.elra.info/mailman3/postorius/lists/corpora.list.elra.info/ To unsubscribe send an email to corpora-leave@list.elra.info