Amaryllo's new patent-pending technology, Voice Print, is a secondary biometric identification tool capable of identifying human voices and differentiating between users. When paired with facial recognition, Voice Print can increase the level of accuracy in human identification.

Here is the voice alert the camera owner will receive when a person has been identified. You will see a box around the captured face, with the identified voice and face labeled. F and V here stand for face and voice, respectively.

Voice Print is currently only available on Athena; it will start the voice recognition process once it detects a human face. That person will be asked to provide 5 to 6 seconds of speech for Athena to pair his or her voice and face. When the process is done, the camera owner will receive the push notification of facial & voice recognition immediately.

The video below demonstrates Voice Print and Face Training.

For voice training: Users need to download the "Easy Voice Recorder" app to record their voices for at least 6 seconds and make sure the recording format is WAV or PCM. User can upload 3 voice files at a time.

For face training: Users can upload 9 photos at a time. Please make sure the photos are in JPG or JPEG format.