Method and system for reducing perplexity in speech recognition via caller identification

• US 5,802,251 A
• Filed: 09/05/1995
• Issued: 09/01/1998
• Est. Priority Date: 12/30/1993
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Lucassen, John M., Sherwin, Elton B. Jr., Vizcaino, Jorge L., Cohen, Paul S.
• Primary Examiner(s): Hafiz, Tariq R.
• Application Number:
  US08/523,755
• Time in Patent Office: 1,092 Days
• Field of Search: 379/38, 379/67, 379/88, 379/96, 379/189, 379/199, 381/41-43, 381/45, 395/2, 395/2.1, 395/2.4, 395/2.5, 395/2.6, 395/2.55, 395/2.62, 395/2.79, 395/2.84
• US Class Current: 704/275
• CPC Class Codes: G10L 15/06 Creation of reference templates; Training of speech recognition systems, e.g. adaptation to the characteristics of the speaker's voice G10L15/14 takes precedence
• CPC Class Codes: G10L 15/065 Adaptation
• CPC Class Codes: G10L 15/20 Speech recognition techniques specially adapted for robustness in adverse environments, e.g. in noise, of stress induced speech G10L21/02 takes precedence
• CPC Class Codes: G10L 17/00 Speaker identification or verification techniques

Abstract

A method and system are disclosed for reducing perplexity in a speech recognition system within a telephonic network based upon determined caller identity. In a speech recognition system which processes input frames of speech against stored templates representing speech, a core library of speech templates is created and stored representing a basic vocabulary of speech. Multiple caller-specific libraries of speech templates are also created and stored, each library containing speech templates which represent a specialized vocabulary and pronunciations for a specific geographic location and a particular individual. Additionally, the caller-specific libraries of speech templates are preferably processed to reflect the reduced bandwidth, transmission channel variations and other signal variations introduced into the system via a telephonic network. The identification of a caller is determined upon connection to the network via standard caller identification circuitry and upon detection of a spoken utterance, that utterance is processed against the core library, if the caller'"'"'s identity cannot be determined, or against a particular caller-specific library, if the caller'"'"'s identity can be determined, thereby greatly enhancing the efficiency and accuracy of speech recognition by the system.

Claims

1. A method for enhancing the accuracy and efficiency of a speech recognition system which processes input frames of speech against stored templates representing speech utilizing a telephonic network, said method comprising the steps of;

creating and storing a core library of speech templates;

creating and storing a plurality of caller-specific libraries of speech templates which each include a vocabulary and pronunciation reflective of a specific geographic location;

attempting to determine an identification of a caller location by utilizing a caller identification system within said telephonic network;

processing an input speech utterance against said core library of speech templates in the event an identification of said caller location within said telephone network is not determined; and

processing an input speech utterance against a particular one of said plurality of caller-specific libraries of speech templates in response to a determination of an identification of said caller location within said telephonic network.

2. The method for enhancing the accuracy and efficiency of a speech recognition system which processes input frames of speech against stored templates representing speech via a telephonic network according to claim 1, wherein said step of creating and storing a plurality of caller-specific libraries of speech templates comprises the step of creating and storing a plurality of caller-specific libraries of speech templates which are processed to reflect variations in each speech utterance which occur as a result of a transmission of each speech utterance within said telephonic network.

3. A system for enhancing the accuracy and efficiency of a speech recognition system which processes input frames of speech against stored templates representing speech utilizing a telephonic network, said system comprising:

means for creating and storing a core library of speech templates;

means for creating and storing a plurality of caller-specific libraries of speech templates which each include a vocabulary and pronunciation reflective of a specific geographical location;

means for attempting to determine an identification of a caller location utilizing a caller identification system within said telephonic network;

means for processing an input speech utterance against said core library of speech templates in the event an identification of said caller location within said telephone network is not determined; and

means for processing an input speech utterance against a particular one of said plurality of caller-specific libraries of speech templates in response to a determination of an identification of said caller location within said telephonic network.

4. The system for enhancing the accuracy and efficiency of a speech recognition system which processes input frames of speech against stored templates representing speech via a telephonic network according to claim 5, wherein said means for creating and storing a plurality of caller-specific libraries of speech templates comprises means for creating and storing a plurality of caller-specific libraries of speech templates which are processed to reflect variations in each speech utterance which occur as a result of a transmission of each speech utterance within said telephonic network.