Audio requirements for imported files

Audio you import into the system must meet certain requirements. Incorrectly formatted audio is not incorporated into the voiceprint File containing a mathematical summary of the vocal patterns of a person’s voice used in the IAFD product to assist with identity verification and detection. model.

Audio property

Requirement

File format

WAV

  • Only files with a valid WAV header can be uploaded.

  • Other file formats, such as MP3, are not supported.

  • Use unencrypted files. Do not use encrypted audio files.

Audio format

16-bit Linear PCM or G.711 is strongly recommended.

  • Use of another audio format is not guaranteed to work for the voiceprint model.

  • 8-bit PCM is not recommended for use due to the poor quality of the audio.

NOTE: Audio in any audio format that was not created by a suite Recorder is unlikely to result in an audio segment that can be used for voiceprint model training due to implementation differences between the audio source and the suite Recorder. Export audio from an external system as a PCM or G.711 WAV file to guarantee compatibility.

Sample rate

8 kHz

Channels

The voice of one person in the channel used for enrollment Process in the IAFD product where an employee, customer, or target is registered and the system is then able to assist with identity verification and watch list detection when that speaker is on a call..

For a mono recording with multiple people speaking, use a third-party editing tool to derive an audio file containing only the voice of a single person. Imported audio is not diarized for enrollment purposes.

Related topics 

Import an audio segment for a voiceprint model