Generating prompts for user vocalisation for biometric speaker recognition

An apparatus for generating a prompt to be vocalised by a user for biometric speaker recognition comprising:

• a prompt element selector for selecting at least one vocal prompt element to form at least part of said prompt from a predetermined set of a plurality of vocal prompt elements;
• wherein the prompt element selector is configured to select the at least one vocal prompt element based on an indication of noise in an audio channel used for the biometric speaker recognition;
• a physical transducer configured to present the at least one vocal prompt element to the user, the at least one vocal prompt element to be vocalised by the user for speaker recognition;
• a microphone for receiving a vocalised prompt,
• wherein the vocalised prompt consists of an audio signal corresponding to sounds of the at least one vocal prompt element as vocalised by the user;
• wherein the apparatus further comprises a speaker recognition module configured to analyse an audio signal corresponding to sounds vocalised by the user in response to the presentation of the at least one vocal prompt element to the user for speaker recognition,
• wherein the speaker recognition module is configured to receive an indication of the selected at least one vocal prompt element and to select at least one user model for an enrolled user based on the selected at least one vocal prompt element, and
• wherein the speaker recognition module is operable in a verification mode to compare said audio signal with the at least one user model for an enrolled user to determine or verify the identity of the user for biometric speaker recognition.

An apparatus for generating a prompt to be vocalised by a user for biometric speaker recognition comprising:

• a prompt element selector for selecting at least one vocal prompt element to form at least part of said prompt from a predetermined set of a plurality of vocal prompt elements;
• wherein the prompt element selector is configured to select the at least one vocal prompt element based on an indication of operating conditions of the biometric speaker recognition and a discrimination score for the at least one vocal prompt element indicative of the likelihood of that at least one vocal prompt element to provide discrimination between users for the current operating conditions;
• a physical transducer configured to present the at least one vocal prompt element to the user, the at least one vocal prompt element to be vocalised by the user for speaker recognition;
• a microphone for receiving a vocalised prompt,
• wherein the vocalised prompt consists of an audio signal corresponding to sounds of the at least one vocal prompt element as vocalised by the user;
• wherein the indication of operating conditions for the biometric speaker recognition comprises one or more of:
• an indication of an acoustic environment in which the at least one vocal prompt element will be vocalised;
• an indication of reverberation in an audio signal derived from a microphone to be used to receive the vocalised prompt;
• an indication of pitch of sounds previously vocalised by the user;
• an indication of a parameter of an acoustic channel to be used to receive the vocalised prompt;
• and
• an indication of vocal characteristics of a user whose identity is to be verified;
• wherein the apparatus further comprises a speaker recognition module configured to analyse an audio signal corresponding to sounds vocalised by the user in response to the presentation of the at least one vocal prompt element to the user for speaker recognition,
• wherein the speaker recognition module is configured to receive an indication of the selected at least one vocal prompt element and to select at least one user model for an enrolled user based on the selected at least one vocal prompt element, and
• wherein the speaker recognition module is operable in a verification mode to compare said audio signal with the at least one user model for an enrolled user to determine or verify the identity of the user for biometric speaker recognition.

An apparatus for generating a prompt to be vocalised by a user for biometric speaker recognition comprising:

• a prompt element selector for selecting at least one vocal prompt element to form at least part of said prompt from a predetermined set of a plurality of vocal prompt elements;
• wherein each of said set of vocal prompt elements provides discrimination between different users above a predetermined threshold level;
• a physical transducer configured to present the at least one vocal prompt element to the user, the at least one vocal prompt element to be vocalised by the user for speaker recognition;
• a microphone for receiving a vocalised prompt,
• wherein the vocalised prompt consists of an audio signal corresponding to sounds of the at least one vocal prompt element as vocalised by the user;
• wherein the apparatus further comprises a speaker recognition module configured to analyse an audio signal corresponding to sounds vocalised by the user in response to the presentation of the at least one vocal prompt element to the user for speaker recognition,
• wherein the speaker recognition module is configured to receive an indication of the selected at least one prompt element and to select at least one user model for an enrolled user based on the selected at least one vocal prompt element, and
• wherein the speaker recognition module is operable in a verification mode to compare said audio signal with the at least one user model for an enrolled user to determine or verify the identity of the user for biometric speaker recognition.