Skip to main content

Voiceprint Models

Voiceprint models are extractors of the biometric characteristics of the voice. They can be used to compute voiceprints.

Version selection: the platform automatically selects the latest versions of verification models.

Model Descriptions

Model NameHandleVersionStatusDescription
Digitaldigitalv1✅ AvailableThe digital voiceprint model is specifically designed and trained for managing higher quality audio, with sample rates of 16KHz or higher. The English and Spanish languages make up a significant part of its training dataset.
Phonephonev1✅ AvailableThe phone voiceprint model is specifically designed and trained for managing phone audio, with sample rates of 8KHz or higher. The telephone channel and the English and Spanish languages make up a significant part of its training dataset.
Defaultdefaultv1⚠️ DeprecatedThe default voiceprint model is deprecated and will be removed in a future platform release.

Model Requirements

Model NameAudio Max DurationAudio Min Voice DurationAudio Min FrequencyVoiceprint Max Total DurationVoiceprint Min Total Voice DurationState
Digital2 minutes4.5 seconds / minimum audios16000 Hz2 minutes4.5 secondsComputable
Phone2 minutes4.5 seconds / minimum audios8000 Hz2 minutes4.5 secondsComputable
Default2 minutes4.5 seconds / minimum audios16000 Hz2 minutes4.5 secondsComputable