PKI 2665 automatically identifies the occurrence of certain keywords and key phrases in audio recordings. PKI 2665’s technology applies state-of-the-art channel compensation techniques that are compatible with the widest possible range of audio sources: GSM/ CDMA, 3G, VoIP, landline, satellite phones, etc. PKI 2665 automatically converts keywords into phonemes. Users can add an unlimited number of keywords and pronunciation variants for each PKI 2655 keyword.
PKI 2665 requires the following input requirements:
Wav or RAW (PCM, 8 or 16 bit, IEEE float 32 bit, A-law or Mu-law, ADPCM) FLAC, OPUS; 8 kHz+ sampling (other audio formats are automatically converted), and the list of keywords.
PKI 2665 uses the following output formats:
XML/ JSON format with all results. In addition, PKI 2665 can provide result files with details of the recognized keywords (keyword, start/end time, path, probability, etc.).
The latest generation of PKI 2665 is about 30x faster than real-time processing on a 1-CPU core. The 4th and older generations of the model are about 10x faster than real-time processing.