Conference call system, method, and computer program product

• US 9,094,526 B2
• Filed: 12/30/2013
• Issued: 07/28/2015
• Est. Priority Date: 11/06/2009
• Status: Active Grant

• Current Assignee: NXP USA, Inc. (NXP Semiconductors NV)
• Original Assignee: Freescale Semiconductor, Inc. (NXP Semiconductors NV)
• Inventors: Krutsch, Robert, Pralea, Radu D.
• Primary Examiner(s): Nguyen, Khai N
• Application Number:
  US14/143,053
• Publication Number:
  US 20140112464A1
• Time in Patent Office: 575 Days
• Field of Search: 379/90.01, 379/93.01, 379/93.21, 379/157, 379/158, 379/201.01, 379/202.01, 379/207.01, 370/259, 370/260, 370/261, 370/262, 455/414.1, 455/416
• US Class Current: 1/1
• CPC Class Codes: H04M 2203/5072 Multiple active speakers
• CPC Class Codes: H04M 2242/06 Lines and connections with preferential service
• CPC Class Codes: H04M 3/566 relating to a participants right to speak arrangements for multi-party communication with floor control, e.g. for conferences, H04L65/4038, H04L65/4046, H04L65/4053
• CPC Class Codes: H04M 3/569 using the instant speaker's algorithm speech detection per se G10L25/78

Abstract

A conference call system comprises an input interface for receiving during a conference call at least two input streams of audio signal, each from another source. A selection unit is connected to the input interface, for selecting a number of dominant speaker streams out of the input streams, the number being less than or equal to a maximum number of dominant speakers value and each of the dominant speaker streams representing speech from a respective dominant speaker. A mixer is connected to the selection unit, for mixing the selected streams into an output stream. The conference call system comprises an output interface for outputting the output stream and a selection control unit connected to the selection unit and the input interface, for dynamically setting, during the conference call, the maximum number of dominant speakers value based on dynamics of the conference call.

Claims

1. A method comprising:

receiving at a conference call system during a conference call a plurality of input speaker streams including a first input speaker stream, each input speaker stream representing an audio signal from a respective audio source;

identifying a subset of the plurality of input speaker streams, wherein each input speaker stream of the subset is designated as a dominant speaker, the total number of input speaker streams in the subset being limited by a maximum number of dominant speakers;

comparing a loudness value of the first input speaker stream to a loudness value of a second input stream of the plurality of input speaker streams that was most recently added to the subset; and

changing the maximum number of dominant speakers based upon the comparison.

2. The method of claim 1, wherein the loudness value of the second input stream is determined at the time that the second input stream was added to the subset.

3. The method of claim 1, wherein the comparing is performed periodically.

4. The method of claim 1, wherein the comparing is performed continuously.

5. The method of claim 1, wherein the comparing is performed in response to detecting sound beyond a threshold at the first input speaker stream.

6. The method of claim 5, wherein detecting sound comprises detecting voice.

7. The method of claim 1, wherein comparing comprises:

determining a loudness value difference between the loudness value of the first input speaker stream and the loudness value of the second input stream;

classifying the loudness value of the first input speaker stream into a first category of a first set of categories;

determining a first weighting factor based on the first category and on the loudness value of the first input speaker stream;

classifying the loudness value difference into in a second category of a second set of categories; and

determining a second weighting factor based on the second category and on the loudness value difference.

8. The method of claim 7, wherein comparing further comprises:

determining, for each rule of a set of rules, an evaluation value based on the first category, the second category, the first weighting factor and the second weighting factor.

9. The method of claim 8, wherein comparing further comprises:

summing the evaluation values; and

comparing the sum of the evaluation values to a threshold value.

10. A conference call system, comprising:

a memory for storing a maximum number of dominant speakers;

an input for receiving during a conference call a plurality of input speaker streams including a first input speaker stream, each input speaker stream representing an audio signal from a respective audio source;

a selection unit operable to;

identify a subset of the plurality of input speaker streams, wherein each input speaker stream of the subset is designated as a dominant speaker, the total number of input speaker streams in the subset being limited by the maximum number of dominant speakers; and

store a count of selected dominant speaker streams as a current number of dominant speakers, the count of selected dominant speaker streams being limited by the maximum number of dominant speakers;

a mixer operable to mix the plurality of input speaker streams into an output stream;

an output interface for outputting said output stream; and

a selection control unit operable to;

compare a loudness value of the first input speaker stream to a loudness value of a second input stream of the plurality of streams that was most recently added to the subset; and

change said maximum number of dominant speakers based upon the comparison.

11. The conference call system of claim 10, wherein the loudness value of the second input stream is determined at the time that the second input stream is added to the subset.

12. The conference call system of claim 10, wherein the selection control unit compares a loudness value periodically.

13. The conference call system of claim 10, wherein the selection control unit compares a loudness value continuously.

14. The conference call system of claim 10, wherein the selection control unit compares a loudness value in response to sound beyond a threshold detected at the first input speaker stream.

15. The conference call system of claim 14, wherein the sound detected is voice.

16. The conference call system of claim 10, wherein compare a loudness value comprises:

determine a loudness value difference between the loudness value of the first input speaker stream and the loudness value of the second input stream;

classify the loudness value of the first input speaker stream into a first category of a first set of categories;

determining a first weighting factor based on the first category and on the loudness value of the first input speaker stream;

classify the loudness value difference into in a second category of a second set of categories; and

determining a second weighting factor based on the second category and on the loudness value difference.

17. The conference call system of claim 16, wherein compare a loudness value further comprises:

determine, for each rule of a set of rules, an evaluation value based on the first category, the second category, the first weighting factor and the second weighting factor.

18. The conference call system of claim 17, wherein compare a loudness value further comprises:

sum the evaluation values; and

compare the sum of the evaluation values to a threshold value.

19. A non-transitory computer readable medium containing a computer program executable by a programming apparatus, said computer program having code portions, when executed by said programming apparatus, perform functions comprising:

receiving at a conference call system during a conference call a plurality of input speaker streams including a first input speaker stream, each input speaker stream representing an audio signal from a respective audio source;

identifying a subset of the plurality of input speaker streams, input speaker streams of the subset being designated as dominant speakers, the total number of input speaker streams in the subset being limited by a maximum number of dominant speakers;

comparing a loudness value of the first input speaker stream to a loudness value of a second input stream of the plurality of input speaker streams that was most recently added to the subset; and

changing the maximum number of dominant speakers based upon the comparison.

20. The computer readable medium of claim 19, wherein the loudness value of the second input stream is determined at the time that the second input stream is added to the subset.