System and method for training diction

• US 8,340,968 B1
• Filed: 01/09/2008
• Issued: 12/25/2012
• Est. Priority Date: 01/09/2008
• Status: Expired due to Fees

• Current Assignee: Lockheed Martin Corporation (Martin Marietta Corporation)
• Original Assignee: Lockheed Martin Corporation (Martin Marietta Corporation)
• Inventors: Gershman, Vladimir
• Primary Examiner(s): RIDER, JUSTIN W
• Application Number:
  US11/971,610
• Time in Patent Office: 1,812 Days
• Field of Search: 704/270, 434/156, 434/157
• US Class Current: 704/270
• CPC Class Codes: G09B 19/06 Foreign languages with audible presentation of material to be studied G09B5/04
• CPC Class Codes: G10L 2015/088 Word spotting

Abstract

A computer-implemented method for automatically training diction of a person acquires a speech data stream of the person as the person is speaking, compares the words in the speech data stream to a set of predefined undesirable phrases provided in a look-up table and upon detection of one of the predefined undesirable phrases in the speech data stream, alerting the person by an alarm.

Claims

1. A computer-implemented method for automatically training diction of a person using an automated system comprising an audio input unit for acquiring the person'"'"'s speech data stream;

a speech processing unit configured with a speech data monitoring program code; and

an alarm unit for generating an alarm signal, said method comprising;

acquiring a speech data stream of the person using the audio input unit as the person is speaking;

identifying words in the speech data stream using the speech processing unit;

comparing the words in the speech data stream to a set of predefined undesirable phrases that are grammatically incorrect provided in a look-up table using the speech processing unit;

generating the alarm signal using the alarm unit upon detection of at least one of the predefined undesirable phrases that are grammatically incorrect from the look-up table in the speech data stream,

storing the number of alarms generated over time, and

generating an output indicative of the progress of the speaker according to either an increase or decrease in the number of alarms generated over time.

2. The method of claim 1, wherein the speech data stream is a digitized data stream.

3. The method of claim 1, wherein alerting the person generating the alarm signal includes generating an audible alarm.

4. The method of claim 1, wherein alerting the person generating the alarm signal includes generating a visible alarm.

5. The method of claim 1, wherein alerting the person generating the alarm signal includes generating a tactile alarm.

6. The method of claim 1, wherein the comparison is independent of the pronunciation of the undesirable phrase by the person speaking.

7. The method of claim 1, further comprising the step of generating an audio playback of a grammatically correct phrase in response to the detection of said at least one of the predefined undesirable phrases in the speech data stream.