Multi-Modal Automation for Human Interactive Skill Assessment

Abstract

A method for screening candidates by a screening entity that includes remotely accessing a screening entity's online web page by a candidate. The candidate inputs responses to a set of queries tailored to the screening entity. The responses thereto are stored in a candidate profile. An interactive audio session is initiated and an interactive information gathering process with the candidate occurs using both the audio session and the computer. Prompts are generated via the audio session and web page. Verbal responses are recorded and confidence scores are automatically calculated. The verbal responses are stored with the candidate profile for later access and analysis by the screening entity. The stored information is analyzed by the screening entity to make quality and criteria judgments about the candidate to determine a hiring potential of the candidate.

We claim: 
     
         1 . A method for screening candidates by a screening entity, which comprises:
 remotely accessing a screening entity's online web page by a potential candidate using an Internet-accessible computer available to the candidate;   using the web page, requesting the candidate to input responses to a set of queries into the computer that are tailored to the screening entity;   storing the candidate's responses to the queries in a candidate profile;   in response to at least one of the candidate's responses, automatically initiating an interactive audio session;   commencing an interactive information gathering process with the candidate using both the interactive audio session and the computer by:
 generating prompts to the candidate via at least one of the interactive audio session and the web page; 
 recording the candidate's verbal responses to the prompts via the interactive audio session; 
 automatically determining a confidence scoring of the recorded verbal responses of the candidate with a computerized speech recognition device by automation; and 
 storing the recorded verbal responses in the candidate profile; 
   storing the candidate profile in a database for later access and analysis in a skill assessment of the candidate by the screening entity; and   analyzing at least one of the candidate profile, the confidence scoring, and the recorded verbal responses by the screening entity to make quality and criteria judgments about the candidate and determine a hiring potential of the candidate.   
     
     
         2 . The method according to  claim 1 , which further comprises carrying out the automatic initiation of the interactive audio session by Voice Over Internet Protocol (VOIP). 
     
     
         3 . The method according to  claim 1 , which further comprises:
 hosting the web page with an external server; and   accessing an internal server that hosts an internal web page of the screening entity.   
     
     
         4 . The method according to  claim 1 , which further comprises one of:
 carrying out the automatic initiation of the interactive audio session step by simultaneously initiating a voice call to the candidate; and   carrying out the automatic initiation of the interactive audio session step by initiating the voice call to the candidate after the candidate has completed responding to the queries.   
     
     
         5 . The method according to  claim 1 , which further comprises initiating the interactive audio session to the candidate with a voice server through a private branch exchange over the telephone network. 
     
     
         6 . The method according to  claim 1 , which further comprises confirming the candidate's identity through a set of identification questions and responses via one of the interactive audio session and the web page. 
     
     
         7 . The method according to  claim 1 , which further comprises carrying out the candidate profile storing step by storing the profile of the candidate one of:
 during the information gathering process; and   after the information gathering process is complete.   
     
     
         8 . The method according to  claim 1 , which further comprises carrying out the profile storing step by storing the profile of the candidate through an entity-secure intranet link. 
     
     
         9 . The method according to  claim 1 , which further comprises carrying out the automatic initiation of the interactive audio session step by initiating the interactive audio session only with outbound calls from the screening entity to, thereby, control access to the information gathering process. 
     
     
         10 . The method according to  claim 1 , which further comprises carrying out the information gathering step by controlling a number and kind of questions presented to the candidate to simulate a real call-center seat. 
     
     
         11 . The method according to  claim 1 , which further comprises carrying out the storing steps one of selectively and continually. 
     
     
         12 . The method according to  claim 1 , wherein the computerized speech recognition device is programmed to parse a particular response into individual words and to either compare the parsed response to a desired responses or to transcribe the parsed response for later use and access by the screening entity. 
     
     
         13 . The method according to  claim 12 , which further comprises carrying out the automated confidence scoring step by judging at least an accuracy and an intelligibility of the candidate's speech utilizing a target phrase represented in a speech recognition grammar. 
     
     
         14 . A method for screening candidates by a screening entity, which comprises:
 initiating a job application procedure by having a candidate remotely connect to a secured online web page of a screening entity using an Internet-accessible computer;   using the secured online web page, querying the candidate to enter candidate-specific personal identification data using the computer;   securely confirming the identification data;   initiating an interactive audio session to the candidate after the identification data is confirmed and, while the candidate is still communicating with the web page and during the interactive audio session, carrying out at least one of the following exercise steps:
 initiating a dictation clarity exercise by instructing candidate to read a script sent to a particular web page viewable by the candidate on the computer and, when finished:
 recording speech from the candidate for subsequent off-line analysis; and 
 automatically determining a confidence scoring of the recorded speech of the candidate with a computerized speech recognition device by automation; and 
 
 initiating a graphic comprehension exercise to screen the candidate's ability to respond to a provided graphic by:
 displaying a graphic to the candidate on the computer; 
 requiring the candidate to answer at least one question regarding the graphic; 
 recording the spoken answers over the interactive audio session; and 
 automatically determining a confidence scoring of each of the spoken answers correlated with how well the recorded phrase matches the expected pronunciation with a computerized speech recognition device by automation; 
 
   after a given number of exercises are complete, storing data recorded for later analysis in a skill assessment of the candidate by the screening entity and informing the candidate that an application process is complete; and   analyzing at least one of the confidence scoring and the recorded responses by the screening entity to make quality and criteria judgments about the candidate and determine the hiring potential of the candidate.   
     
     
         15 . The method according to  claim 14 , which further comprises securely confirming the identification data. 
     
     
         16 . The method according to  claim 14 , which further comprises carrying out the secure confirming step by one of directly with the candidate and indirectly through an email sent to the candidate's email address that requires a response from the candidate. 
     
     
         17 . The method according to  claim 14 , which further comprises carrying out the interactive audio session initiating step with a voice server immediately after the identification data is confirmed. 
     
     
         18 . The method according to  claim 14 , which further comprises carrying out at least one of the following exercise steps:
 initiating a voice quality exercise over the interactive audio session by asking the candidate to repeat spoken phrases over the interactive audio session, recording the candidate's speech for each phrase for subsequent off-line analysis, and automatically scoring each recorded phrase with a confidence level correlated with how well the recorded phrase matches an expected pronunciation;   initiating a question-answer exercise by transmitting specific multiple-choice questions over the interactive audio session, prompting the candidate to answer each question after it is presented, recording the candidate's speech for each answer, and automatically scoring each recorded answer with a confidence level correlated to how well the recorded phrase matches an expected pronunciation;   initiating a service knowledge exercise to determine the candidate's ability to comprehend an example of what good customer service is and whether the candidate can intelligently describe such an experience by having the candidate give a speech regarding such an experience, recording the speech for subsequent off-line analysis, and applying a subjective confidence scoring to the recorded speech; and   initiating a speaking satisfaction exercise with the candidate to evaluate the kind of experience that a potential customer will have after speaking to the candidate by giving the candidate a random one of a series of predefined situations in which candidate is required to role-play an operator answering in real-time a call from a virtual customer, recording the conversation between the candidate and the virtual customer for subsequent off-line analysis, and applying a subjective confidence scoring to the recorded conversation.   
     
     
         19 . The method according to  claim 18 , which further comprises carrying out the exercise step by storing the candidate's recorded speech for subsequent off-line analysis. 
     
     
         20 . The method according to  claim 18 , which further comprises carrying out the confidence scoring at least one of automatically and with human judgment. 
     
     
         21 . The method according to  claim 18 , which further comprises carrying out the question-answer exercise step by transmitting the multiple-choice questions by at least one of a human voice and a machine voice. 
     
     
         22 . The method according to  claim 14 , which further comprises carrying out the graphic comprehension exercise step by storing the recorded spoken answers for subsequent off-line analysis. 
     
     
         23 . The method according to  claim 14 , which further comprises carrying out the graphic comprehension exercise step by providing a map as the graphic and screening the candidate's ability to give driving directions. 
     
     
         24 . The method according to  claim 18 , which further comprises carrying out the speaking satisfaction exercise step by storing the conversation for subsequent off-line analysis. 
     
     
         25 . The method according to  claim 14 , which further comprises providing the candidate with at least one of:
 a timeframe for hearing from the screening entity; and   a call number and a date for checking on the candidate's application.