Methods and apparatus for identifying unknown speakers using a hierarchical tree structure

A system for identifying a speaker in an audio source, comprising:

• a memory that stores computer-readable code;
• and
• a processor operatively coupled to said memory, said processor configured to implement said computer-readable code, said computer-readable code configured to:
• generate a plurality of speaker models each corresponding to a different speaker, said speaker models including at least one background model corresponding to an unenrolled speaker, said speaker models containing a collection of distributions of audio feature data associated with that speaker;
• merge similar speaker models on a layer by layer basis so as to generate a hierarchical speaker model tree;
• identify potential segment boundaries in said audio source;
• and
• assign a speaker label to each identified segment, said speaker label being selected by comparing said identified segment to said speaker models.

An article of manufacture, comprising:

• a computer readable medium having computer readable code means embodied thereon, said computer readable program code means comprising:
• a step to generate a plurality of speaker models each corresponding to a different speaker, said speaker models including at least one background model corresponding to an unenrolled speaker, said speaker models containing a collection of distributions of audio feature data associated with that speaker;
• merge similar speaker models on a layer by layer basis so as to generate a hierarchical speaker model tree;
• identify potential segment boundaries in said audio source;
• and
• assign a speaker label to each identified segment, said speaker label being selected by comparing said identified segment to said speaker models.

A method for identifying a speaker in an audio source, said method comprising the steps of:

• generating a plurality of speaker models each corresponding to a different speaker, said speaker models including at least one background model corresponding to an unenrolled speaker, said speaker models containing a collection of distributions of audio feature data associated with that speaker;
• merging similar speaker models on a layer by layer basis so as to generate a hierarchical speaker model tree;
• identifying potential segment boundaries in said audio source;
• assigning a speaker label to each identified segment, said speaker label being selected by comparing said identified segment to said speaker models;
• presenting said user with said assigned speaker labels;
• and
• prompting a user for an identity of an unenrolled speaker.

A system for identifying a speaker in an audio source, comprising:

• a memory that stores computer-readable code;
• and
• a processor operatively coupled to said memory, said processor configured to implement said computer-readable code, said computer-readable code configured to:
• generate a plurality of speaker models each corresponding to a different speaker, said speaker models including at least one background model corresponding to an unenrolled speaker, said speaker models containing a collection of distributions of audio feature data associated with that speaker;
• merge similar speaker models on a layer by layer basis so as to generate a hierarchical speaker model tree;
• identify potential segment boundaries in said audio source;
• assign a speaker label to each identified segment, said speaker label being selected by comparing said identified segment to said speaker models;
• present said user with said assigned speaker labels;
• and
• prompt a user for an identity of an unenrolled speaker.

An article of manufacture, comprising:

• a computer readable medium having computer readable code means embodied thereon, said computer readable program code means comprising:
• a step to generate a plurality of speaker models each corresponding to a different speaker, said speaker models including at least one background model corresponding to an unenrolled speaker, said speaker models containing a collection of distributions of audio feature data associated with that speaker;
• a step to merge similar speaker models on a layer by layer basis so as to generate a hierarchical speaker model tree;
• a step to identify potential segment boundaries in said audio source;
• a step to assign a speaker label to each identified segment, said speaker label being selected by comparing said identified segment to said speaker models;
• a step to present said user with said assigned speaker labels;
• and
• a step to prompt a user for an identity of an unenrolled speaker.