Spanish recognition – accented words

[chickenthief2]

All of my accented words seem to be using the fallback method. Am I doing something wrong?

For example, ESTÁS will appear ‘in the normalized array’ as “EST\U00c1S”,

then…

“The word ESTÁS was not found in the dictionary …./AcousticModelSpanish.bundle/LanguageModelGeneratorLookupList.text/LanguageModelGeneratorLookupList.text”

Also, other words – without accents – seem to be missing from the Spanish dictionary. Is the Spanish dictionary just smaller than the English one?

Thanks a lot!

[Halle Winkler]

Is this from an NSArray or a text file?

[chickenthief2]

Yeah, I thought that might be it, but I’ve tried both ways.

[Halle Winkler]

Is it with all accented letters or just specific ones?

[chickenthief2]

I believe it’s all of them but I’ll double check on that when I get the chance. Words without accents that are in the dictionary seem to work Ok.

I’ll also note that I have not made any modifications to the dictionary.

Thanks

[Halle Winkler]

OK, let me know device and iOS and show me an NSArray of words that causes the result and I will take a look at it.

[chickenthief2]

7.0.6 and iPhone 5s

ESTÁS NOT found (fallback method):

NSArray *words = [NSArray arrayWithObjects: @”HOLA”,
@”BUENOS DÍAS”,
@”BUENAS TARDES”,
@”BIENVENIDO”,
@”CÓMO ESTÁS”,
@”LOS ESTADOS UNIDOS”,nil

ESTÁS found:

NSArray *words = [NSArray arrayWithObjects: @”HOLA”,
@”BUENOS DÍAS”,
@”BUENAS TARDES”,
@”BIENVENIDO”,
@”CÓMO ESTÁS”,
,nil

[Halle Winkler]

In this case, are all the other accented characters found?

[chickenthief2]

yes

[Halle Winkler]

This was due to the lookup dictionary being sorted differently than LanguageModelGenerator expects and has been fixed for the next release of OpenEars.

[Author]

Posts