OK, in order to perform recognition on an existing WAV file, you can submit a WAV file that is 16-bit/16k/mono to this method of OEPocketsphinxController:
- (void) runRecognitionOnWavFileAtPath:(NSString *)wavPath usingLanguageModelAtPath:(NSString *)languageModelPath dictionaryAtPath:(NSString *)dictionaryPath acousticModelAtPath:(NSString *)acousticModelPath languageModelIsJSGF:(BOOL)languageModelIsJSGF;
and SaveThatWave creates files in that format. But, the method above won't create a different recognition than the main OpenEars recognition method creates. Does that answer help?