El 14/05/2010 02:00 a.m., Pierre Abbat escribió: > On Thursday 13 May 2010 23:48:17 Leo Molas wrote: >> coi rodo >> >> I've been working on a project for a couple of weeks, mostly for myself, >> but it might be useful for the community. It's a dictionary, created in >> LaTeX. > > Looks good! > >> TODO: >> - I want to put the IPA representation in each word, since I can do it >> automatically. > > Are you going to use the same allophone for all occurrences of a letter? That > would seem pointless. Some phonemes have allophones, and I at least don't > pronounce them the same way all the time: > "n" can be [n] or [ŋ]. I usually pronounce it [ŋ] before "x", "g", and "k", > and [n] elsewhere, but for some words such as "tcunCIN" I make an exception. > "r" can be any of several rhotics. I usually tap or roll it except when it's > consonantal, in which case it's like English "burble", or preceded by "x", in > which case both sounds come out as simultaneous uvular and alveolar trills > distinguished only by voice. I've met people at jbovlaste who don't roll > their r's. > "'" can be [h] or [θ]. I say [h] in all words except sometimes "fu'arka" > (and "fu'orka", which is a variant thereof). I've heard that some people use > [θ] generally, but I don't remember if I've met them. > > Pierre The idea was to automatically create the IPA representation with the most common pronuntiations, and have a chapter in the book explaining how to pronounce lojban (like "'" can be [h] or [θ], or in long consonants cluster you can put "i"'s in the middle). Anyway, the main idea of showing this to you was getting feedback, and if you say it's pointless, fair enough. I actually don't need IPA; it was just to make the dictionary homogeneous to others in others languages. -- My lojban journal: http://learninglojban.wordpress.com My personal blog: http://leomolas.tumblr.com
Attachment:
signature.asc
Description: OpenPGP digital signature