Apparatus and method for providing voice recognition for multiple speakers

• US 7,844,454 B2
• Filed: 03/18/2003
• Issued: 11/30/2010
• Est. Priority Date: 03/18/2003
• Status: Active Grant

• Current Assignee: Arlington Technologies, LLC (Dominion Harbor Enterprises, LLC)
• Original Assignee: Avaya Incorporated
• Inventors: Gentle, Christopher Reon, Orbach, Julian James, Harrison, Rodney, Coles, Scott David, Goringe, Christopher Michael
• Primary Examiner(s): Chawan; Vijay B
• Application Number:
  US10/393,729
• Publication Number:
  US 20040186712A1
• Time in Patent Office: 2,814 Days
• Field of Search: 704/235, 704/246, 704/231
• US Class Current: 704/235
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence

Abstract

Utilizing individual voice recognition units for each speaker in a conference to perform automatic transcription of that speaker'"'"'s contribution to the conference. The output of each of the voice recognition units is then merged on a time basis to produce a textual transcription of the entire telecommunication conference call.

Claims

1. A method for converting all audio information from a plurality of participants to combined textual information, comprising the steps of:

performing independent voice recognition operations on audio information from each of the plurality of participants by a separate voice recognition unit for each of the plurality of participants thereby creating a stream of textual information for each of the plurality of participants;

inserting a time stamp with each word of each stream of textual information; and

combining the streams of textual information and time stamps into combined textual information on the basis of the time stamps in each of the plurality of streams of textual information whereby the combined textual information is textual information of the all audio information.

2. The method of claim 1 comprises the step of training each voice recognition unit for its participant.

3. The method of claim 1 wherein the step of combining comprises the step of sorting the words from each of the streams into the combined information on the basis of the time stamps.

4. The method of claim 1 further comprises storing the audio information with time stamps.

5. A processor-readable medium comprising processor-executable instructions configured for:

performing independent voice recognition operations on audio information from each of the plurality of participants by a separate voice recognition unit for each of the plurality of participants thereby creating a stream of textual information for each of the plurality of participants;

inserting a time stamp with each word of each stream of textual information; and

combining the streams of textual information and time stamps into combined textual information on the basis of the time stamps in each of the plurality of streams of textual information whereby the combined textual information is textual information of the audio.

6. The processor-readable medium of claim 5 comprises the training each voice recognition unit for its participant.

7. The processor-readable medium of claim 5 wherein the combining comprises the sorting the words from each of the streams into the combined information on the basis of the time stamps.

8. The processor-readable medium of claim 5 further comprises storing the audio information with time stamps.

9. An apparatus for converting all audio information from a plurality of participants to combined information, comprising:

a plurality of voice recognition units each implemented by an individual processor and each assigned to one of the plurality of participants whereby audio information from each of the assigned participant is processed to produce a textual stream with a time stamp for each word of each stream of textual information; and

a controller response to the textual streams with time stamps for combining the streams into combined textual information.