Apparatus and methods for managing resources for a system using voice recognition

• US 9,685,154 B2
• Filed: 09/24/2013
• Issued: 06/20/2017
• Est. Priority Date: 09/25/2012
• Status: Active Grant

• Current Assignee: nVoq, Inc.
• Original Assignee: nVoq, Inc.
• Inventors: Corfield, Charles
• Primary Examiner(s): ZHU, RICHARD Z
• Application Number:
  US14/035,641
• Publication Number:
  US 20140088962A1
• Time in Patent Office: 1,365 Days
• CPC Class Codes: G10L 15/01 Assessment or evaluation of speech recognition systems
• CPC Class Codes: G10L 15/063 Training
• CPC Class Codes: G10L 15/065 Adaptation
• CPC Class Codes: G10L 15/07 to the speaker
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence

Abstract

The technology of the present application provides a method and apparatus to managing resources for a system using voice recognition. The method and apparatus includes maintaining a database of historical data regarding a plurality of users. The historical database maintains data regarding the training resources required for users to achieve an accuracy score using voice recognition. A resource calculation module determines from the historical data an expected amount of training resources necessary to train a new user to the accuracy score.

Claims

1. A method performed on at least one processor for determining training time for a new user to achieve a predetermined accuracy score, the method comprising the steps of:

performing one or more training exercises for each of a plurality of users of a speech recognition engine, wherein the one or more training exercises comprise the steps of;

transmitting known text to each of the plurality of users;

receiving at least one audio file from each of the plurality of users, wherein the at least one audio file from each of the plurality of users corresponds to the known text;

transcribing each of the at least one audio file from each of the plurality of users using the speech recognition engine to generate at least one transcribed audio file;

correcting each of the at least one transcribed audio file using the known text transmitted to each of the plurality of users; and

based at least in part on the corrections, determining an accuracy score for each of the at least one transcribed audio file based at least in part on the corrections associated with each of the plurality of users;

storing historical training data associated with the plurality of users, wherein the historical training data includes a first set of data indicative of an amount of time for the training exercises performed with each user, a second set of data indicative of a number of training exercises performed with each user, and a third set of data indicative of the accuracy score associated with each user;

based at least in part on the stored historical training data, calculating, based at least in part on the stored historical training data, an estimated amount of time for the speech recognition engine to achieve a predetermined accuracy score for a new user,

estimating, by the at least one processor, an attrition rate of current employees that use the speech recognition engine; and

determining, by the at least one processor, a hiring window for at least one new employee, wherein the hiring window is in advance of attrition of at least one current employee and is based at least in part on the estimate of the amount of time for the speech recognition engine to achieve the predetermined accuracy score, such that the speech recognition engine is estimated to achieve the predetermined accuracy score for the new employee prior to a departure of the at least one current employee.