Sign in

v2

latestSwagger 2.02026-07-1371728.8 KB
text-independent

Enroll Profile

Adds an enrollment to existing profile. If the minimum number of requested enrollment audios is reached, a voice print is created. If the voice print was created before, it gets recreated from all existing enrollment audios including the new one.

Limitations:

  • Minimum audio input length per request is 1 second

  • Maximum audio input length per request is 120 seconds

  • Minimum total effective speech length (excluding silence and other non-speech frames) for creating a voiceprint is 20 seconds This limitation can be disabled by setting ignoreMinLength to true.

  • Maximum total audio input length allowed for creating a voiceprint is 300 seconds

  • Minimum audio Signal-to-noise ratio (SNR) is 0dB

post/text-independent/profiles/{profileId}/enrollments

Query parameters

ignoreMinLengthboolean

If true, a voice print will be created immediately for this profile regardless of how much speech is supplied or stored. Default is false.

Response

Created

profileIdstring uuid

Unique identifier for profile id (guid).

enrollmentStatus'Enrolling' | 'Training' | 'Enrolled'

Status representing the current state of the profile. Available values are:

  • Enrolling: profile has no voice print and not ready for recognition requests.
  • Training: voice print of profile is being created and can’t be used for recognition at the moment.
  • Enrolled: profile has a voice print and ready for recognition requests.
enrollmentsCountinteger

Number of enrolment audios accepted for this profile.

enrollmentsLengthnumber

Total length of enrollment audios accepted for this profile in seconds.

enrollmentsSpeechLengthnumber

Summation of pure speech (which is the amount of audio after removing silence and non-speech segments) across all profile enrollments in seconds.

remainingEnrollmentsSpeechLengthnumber

Amount of pure speech (which is the amount of audio after removing silence and non-speech segments) needed to complete profile enrollment in seconds.

audioLengthnumber

This enrolment audio length in seconds.

audioSpeechLengthnumber

This enrollment audio pure speech (which is the amount of audio after removing silence and non-speech segments) length in seconds.

Example response

{
  "profileId": "49a36324-fc4b-4387-aa06-090cfbf0064f",
  "enrollmentStatus": "Enrolling",
  "enrollmentsCount": 1,
  "enrollmentsLength": 1.83,
  "enrollmentsSpeechLength": 1.35,
  "remainingEnrollmentsSpeechLength": 18.65,
  "audioLength": 1.83,
  "audioSpeechLength": 1.35
}