IDOL Speech Server requires a classifier file for each language that you want to identify. The base classifier pack provided with IDOL Speech Server covers the following languages for broadband (16 kHz) audio; for any other languages, or to process telephony (8 kHz) audio, you must create your own classifiers.
Arabic | Hebrew | Portuguese |
Danish | Italian | Romanian |
Dutch | Japanese | Russian |
English | Korean | Slovak |
French | Mandarin | Spanish |
German | Persian | Swedish |
Greek | Polish |
For each language, gather a set of audio files that contain speech from that language only. The audio must not include excessive music or noise, although it must closely resemble the quality of audio that you intend to process.
You must convert each audio file into a language identification feature (.lif) file.
|