SPEAKER RECOGNITION IN A MULTI-SPEAKER ENVIRONMENT AND COMPARISON OF SEVERAL VOICE PRINTS TO MANY

摘要

One-to-many comparisons of callers' voice prints with known voice prints toidentify any matches between them. When a customer communicates with aparticular entity, such as a customer service center, the system makes arecording of the real-time call including both the customer's and agent'svoices. The system segments the recording to extract at least a portion of thecustomer's voice to create a customer voice print, and it formats thesegmented voice print for network transmission to a server. The servercompares the customer's voice print with multiple known voice prints todetermine any matches, meaning that the customer's voice print and one of theknown voice prints are likely from the same person. The identification of anymatches can be used for a variety of purposes, such as determining whether toauthorize a transaction requested by the customer.