This invention relates to a conference call system and method, as well as a computer program product.
In current voice conferencing systems, a speaker selection algorithm in a conferencing bridge detects active speakers and creates an output stream by mixing the audio for the active speakers or active participants. The active stream is then communicated to the participants on the conference call. However, selection of the active speakers involves selecting a predetermined number of most active speakers, commonly referred to as the dominant speakers, based on energy levels of voice communications received from the telephony endpoints where the active speakers are located. All other speakers are excluded from the speaker selection algorithm when speech from the dominant speakers is received.
For example, United States Patent Application publication US20071263821 A1 describes a method and apparatus to provide speaker selection in a multi-party conference call. The method comprises processing a speaker queue for at least one new speaker and monitoring when a number of dominant speakers is less than a predetermined number of dominant speakers. When the predetermined number of dominant speakers is less than the predetermined number, the method automatically, without human intervention, adds the new speaker from the speaker queue to the dominant speakers.
Conventional speaker selection algorithms by design end up not allowing new speakers to join until one of the dominant speakers has been quiet for a while. Although this eliminates interruptions it also precludes new speakers from the opportunity to speak if the dominant speakers continue to keep speaking
The present invention provides a conference call system, a conference call method, and a computer program product as described in the accompanying claims.
Specific embodiments of the invention are set forth in the dependent claims.
These and other aspects of the invention will be apparent from and elucidated with reference to the embodiments described hereinafter.
Further details, aspects and embodiments of the invention will be described, by way of example only, with reference to the drawings. In the drawings, like reference numbers are used to identify like or functionally similar elements. Elements in the figures are illustrated for simplicity and clarity and have not necessarily been drawn to scale.
Because the illustrated embodiments of the present invention may for the most part, be implemented using electronic components and circuits known to those skilled in the art, details will not be explained in any greater extent than that considered necessary for the understanding and appreciation of the underlying concepts of the present invention and in order not to obfuscate or distract from the teachings of the present invention.
Referring to
The input streams each comprise data or signals representing at least audio from the source, and may in addition thereto comprise video and/or data. The audio may for example consist of voiced audio, e.g. speech, and comprise other types of audio as well, such as background noise, non-voiced sounds, background rumour. The audio may for example be unfiltered or have been filtered to remove undesired components, such as noise, non-voiced sounds, background rumour or otherwise.
The conference call system 1 shown in
The selection may be performed using any criteria suitable to select dominant speaker streams. For example, the selection may be performed automatically, without human interference by selecting the Kmax number of loudest streams. In this respect, the term “loudest” refers to the streams in which the highest intensity in a signal of interest is observed. The signal of interest may for example be the audio signal as a whole or a part of the audio signal, such as the voiced part thereof, without background noise and non-voiced parts. The signal of interest may for example be taken over one or more periods of time, e.g. the loudness during one or a multiple of time-windows may be used as a selection criterion.
For instance, a loudness number A may be calculated for the audio in each stream and the streams with the Kmax highest loudness numbers λ1, λ2, . . . , λKmax may be selected. A suitable method for calculating the loudness number is described in: Prasad, Venkatesha R.; Kuri, Joy; Jamadagni, H S; Dagale, Haresh; and Ravindranath, Ravi A., “Automatic Addition and Deletion of Clients in VoIP Conferencing,” ISCC, pp. 0386, Sixth IEEE Symposium on Computers and Communications (ISCC'01), 2001, hereinafter referred to as “the ISCC 2001 paper” and incorporated herein by reference.
The loudness number λ may for example be a function of the present and past amplitudes of the input stream. The loudness number λi of a stream i can for example be computed from the current activity L1 during a “recent past” window Wrp, the past activity L2 in a “distant past” window Wdp and the overall past activity L.sub.3 in a “overall past” window wop. The current activity L.sub.1 may for example be the moving average of the amplitude during the recent past window Wrp. The past activity L2 may for example be the moving average of the amplitude during the distant past window Wdp. The past activity L2 may for example be the moving average of the amplitude during the distant past window Wdp. The overall past activity L3 may for example be the moving average of the amplitude during the distant past window Wap.
As illustrated in
The activities L1, L2, L3 may for example be calculated by performing an operation as can be described by the mathematical formula:
in which Xk,i is the amplitude of the k-th sample of the i-th input stream Θ is a binary operation which operates relative to the maximum amplitude max(X(k)) found in the input streams and may for example be calculated by performing an operation as can be described by the mathematical formula:
in this equation (2), the value 0.1 is a threshold value, which may be set to any value suitable for the specific implementation and for example be selected from the group of 0.1 and 0.2, as well as all intermediate points between 0.1 and 0.2.
The loudness number λ may for example be a weighted sum of the activities L.sub.1, L.sub.2, L.sub.3, and for example be calculated by performing an operation as can be described by the mathematical formula:
λ=α1*L1+α2*L2+α3*L3
where α1, α2, α3 are constants with a suitable value. Values suitable have found to be, for instance, α1=0.6; α2=0.3; α3=0.3 but other values may be used as well.
The dominant speaker streams may be selected in another manner, and for example be selected based on the content of the speech in the input streams, e.g. in the manner described in US 2007/263821 or any other suitable manner.
In the example shown in
In the shown example, the output interface 117 splits the output stream 110 into multiple streams 111-113. However, alternatively, the output interface 117 may output the output stream 110 which is then distributed further to the respect communication units by another node in a telecommunication network.
In the example of
The selection control unit 18 may be implemented to set the maximum number of dominant speakers value Kmax based on dynamics of the conference in any manner suitable for the specific implementation. For example, the selection control unit 18 may set the maximum number of dominant speakers value Kmax to equal the number of input streams of which the loudness exceeds a predetermined threshold or any other suitable manner.
Referring to
The comparator 181 can compare the input streams 101-103 and select the dominant speaker streams out of the input streams 101-103. A suitable technique for selecting the dominant speaker streams is described in Prasad, Venkatesha R.; Jamadagni, H S; and Shankar, H N: “On the Problem of Specifying the Number of Floors for a Voice-Only Conference on Packet Networks”, International Conference on Information Technology: Research and Education, 2003. ITRE2003, 11-13 August, New Jersey, USA, 22-26, hereinafter referred to as “the ITRE2003 paper”, incorporated herein by reference. The comparator 181 may for example select the inputs streams for which the loudness number λ exceed a predetermined threshold representative for dominant speaker streams, up to a maximum of Kmax. If the number of input streams above the threshold exceeds Kmax as explained above, the comparator 181 may for example compare the input streams to select the Kmax number of input streams with the highest loudness numbers. The comparator 181 may periodically repeat the comparison, and may thereby avoid that a dominant stream remains selected even when the corresponding input stream does not contain voice any more.
In the shown example, the comparator 181 is connected to the input interface 14 via respective calculators 188-190 which are arranged to calculate for each of the input streams 101-103 the loudness value λ from a parameter of the respective input stream representative of the loudness of the audio, e.g. as explained above. In the shown example, a separate calculator is present for each input stream 101-103 and the calculators 188-190 are connected to the input interface 14. For each of the calculators 188-190 a voice detector 185-187 is connected with an input to the input interface 14. The voice detectors 185-187 can detect voice in the input streams 101-103. The voice detectors 185-187 enable the respective calculator 188-190 when voice is detected and disable the respective calculator 188-190 or otherwise ensure that it outputs the lowest loudness number when no voice is detected. Thereby, only those input streams from which voiced signals are receive can potentially be selected by the comparator 181. The voice detectors 185-187 may be implemented in any manner suitable to detect voice, such as known in the art, and for the sake of brevity are not described in further detail. A suitable implementation is described in for example Prasad, Venkatesha R.; Sangwan, Abhijeet; Jamadagni, H S; Chiranth, M C and Sah, Rahul “Comparison of Voice Activity Detection Algorithms for VoIP”, Seventh International Symposium on Computers and Communications, 2002. ISCC 2002, 1-4 July, Taormina-Giardini Naxos, Italy, pp. 530-535, hereinafter referred to as the “ISCC 2002 paper” and incorporated herein by reference.
The selection control unit 18 may for instance comprise a first logic unit 301 and a second logic unit 302. In the shown example, the logic units 301,302 are connected with their respective inputs to the interface 14. The logic units 301,302 are connected with their outputs to the memory unit 303 in which the maximum number of dominant speaker value Kmax is stored.
As explained below in more detail with reference to
The first logic unit 301 may for example be connected to a new speaker detector 180. In the shown example, the new speaker detector connects the first logic unit 301 to the input interface 14. However, it will be apparent that other units may be present between the first logic unit 301 and the input interface 14.
The new dominant speaker detector 180 can detect out of the input streams additional dominant speaker streams, i.e. additional to the input streams selected as dominant speaker streams. If the detectors 180 detects an additional dominant speaker stream, the detector 180 outputs an additional dominant speaker notification to the first logic unit 301. The first logic unit 301 then determines, e.g. in response to the notification, whether or not to increase the maximum number of dominant speakers value. For example, when the current number of dominant speakers is below the maximum number of dominant speakers value Kmax the first logic unit 301 may maintain the maximum number of dominant speakers value as is, and when the current number of dominant speakers is equal to the maximum number of dominant speakers value, the first logic unit 301 may increment the value Kmax by 1.
The new dominant speaker detector 180 can detect additional dominant speaker streams in any manner suitable for the specific implementation. To that end, the new dominant speaker detector 180 can compare one or more parameters of the input stream with one or more criteria suitable to detect additional dominant speaker streams.
The parameter may for example be the loudness. Referring to the example shown in
In the shown example, the new dominant speaker detector 180 further includes voice detectors 185-187 connected with their inputs to the detector inputs 182-183 and with their outputs to respective control inputs of the calculators 188-190. The voice detectors 185-187 compare one or more parameters of the input streams with criteria suitable to detect voice in the input streams. The new dominant speaker detectors 180 enable a corresponding calculator when voice is detected in an input stream which before did not comprise voice. Thereby, it may be ensured that only input streams which comprise voice can be assigned as dominant speakers and that, for example, input streams with a high audio amplitude due to noise or non-voiced audio (e.g. music) remain non-dominant and accordingly, the overall quality of the conference call can be improved. Although in the example of
The shown new dominant speaker detector 180 further comprises a classifier 191 connected to the calculator. The classifier 191 classifies a non-dominant stream of the input streams 101-103 based on at least the loudness value corresponding to the respective non-dominant stream i in one or more categories M1-M6,N1-N7. In the shown example, the classifier 191 classifies the input stream in respective categories of two sets M1-M6 and N1-N7. The classification in the first set M1-M6 is based on the loudness value λ, whereas the classification in the second set N1-N7 is based on the differential loudness value Δλi which is the difference between the loudness value Δλi and the loudness value Δλa of the last dominant speaker elected at the point the last dominant speaker was elected.
For example, for a loudness value λi of T2, the input stream i will be classified in category M2 with WM=1. For a loudness value λi of T3, the input stream will be classified in category M3 with WM=1. For a loudness value λi between T2 and T3, e.g. λi=A, the input stream will be classified in categories M2 and M3 with a weighing factor Wm2(A) for M2 as defined by the continuous line and a weighing factor Wm3(A) for M3 as defined by the dashed line.
The shown new dominant speaker detector 180 further comprises a detector logic unit 192 connected to the classifier 191. When in operation, the detector logic unit 192 evaluates one, or more than one, rule based on the classification and determines for each of the rules an evaluation value. For example, the detector logic unit 192 may evaluate the following rules R0-R19, where i refers to input stream i:
R0: IF (iεM2 & iεN1) then D=1 OR
R1: IF (iεM2 & iεN2) then D=1 OR
R2: IF (iεM2 & iεN3) then D=1 OR
R3: IF (iεM3 & iεN1) then D=1 OR
R4: IF (iεM3 & iεN2) then D=1 OR
R5: IF (iεM3 & iεN3) then D=1 OR
R6: IF (iεM3 & iεN4) then D=1 OR
R7: IF (iεM3 & iεN5) then D=1 OR
R8: IF (iεM4 & iεN1) then D=1 OR
R9: IF (iεM4 & iεN2) then D=1 OR
R10: IF (iεM4 & iεN3) then D=1 OR
R11: IF (iεM4 & iεN4) then D=1 OR
R12: IF (iεM4 & iεN5) then D=1 OR
R13: IF (iεM5 & iεN1) then D=1 OR
R14: IF (iεM5 & iεN2) then D=1 OR
R15: IF (iεM5 & iεN3) then D=1 OR
R16: IF (iεM5 & iεN4) then D=1 OR
R17: IF (iεM5 & iεN5) then D=1 OR
R18: IF (iεM5 & iεN6) then D=1 OR
R19: IF iεM6 then D=1.
The detector logic unit 192 may for each of these rules calculate an evaluation value E, for example by performing a calculation as can be described by Eq=Dq*(WM(q)+WN(q), in which WM(q) and WN(q) represent the weighing factors for the categories Mq and Nq used in the rule q. For example for R18, supposing that as illustrated in
The shown new dominant speaker detector 180 further comprises an adder 193 connected to the detector logic unit 192 and a comparator 194 connected to the adder 193. The adder 193 receives from the detector logic unit 192 the evaluation values and adds the evaluation values E1, E2, . . . , Ej to obtain a summed value Σ. The comparator 194 then compares the summed value Σ with a threshold value Tr and outputs, at the detector output 195, a maximum number of dominant speakers value increase notification when the summed value Σ exceeds the threshold value Tr.
Referring to back to
In the example of
The second logic unit 302 further comprises a loudness comparator 211 connected to the input interface 14. The loudness comparator 211 compares an intensity of at least voice in one, or more than one, of the dominant speaker streams with a loudness threshold value. The example of
When in one, or more than one, of the dominant speaker streams no voice is detected and/or when for one, or more than one, of the dominant speaker streams the intensity is determined by the loudness comparator 211 to be below the loudness threshold value, Tr, the decrement evaluation unit 200 outputs the notification to the second logic unit 302. In response to the notification, the second logic unit 302 then decreases the maximum number of dominant speakers value Kmax. For generating the notification, the shown example comprises an OR-gate 213 which is connected with a first input port to the comparator 211 and with a second input port to the voice detectors. The first input port is asserted by the loudness comparator 211 when one, or more than one, of the dominant speaker streams the intensity is determined by the loudness comparator 211 to be below the loudness threshold value, Tr. The second input port is asserted when one or more of the voice detectors 204-206 does not detect voice. The OR-gate 213 outputs a notification in the form of a binary 1 when at least one of the input ports is asserted, and outputs nothing (in the form of a binary 0) when none of the input ports is asserted.
The decrement evaluation unit 200 shown in
The invention may also be implemented in a computer program for running on a computer system, at least including code portions for performing steps of a method according to the invention when run on a programmable apparatus, such as a computer system or enabling a programmable apparatus to perform functions of a device or system according to the invention.
A computer program is a list of instructions such as a particular application program and/or an operating system. The computer program may for instance include one or more of: a subroutine, a function, a procedure, an object method, an object implementation, an executable application, an applet, a servlet, a source code, an object code, a shared library/dynamic load library and/or other sequence of instructions designed for execution on a computer system.
The computer program may be stored internally on computer readable storage medium, e.g. as shown in
A computer process typically includes an executing (running) program or portion of a program, current program values and state information, and the resources used by the operating system to manage the execution of the process. An operating system (OS) is the software that manages the sharing of the resources of a computer and provides programmers with an interface used to access those resources. An operating system processes system data and user input, and responds by allocating and managing tasks and internal system resources as a service to users and programs of the system.
The computer system may for instance include at least one processing unit, associated memory and a number of input/output (I/O) devices. When executing the computer program, the computer system processes information according to the computer program and produces resultant output information via I/O devices.
In the foregoing specification, the invention has been described with reference to specific examples of embodiments of the invention. It will, however, be evident that various modifications and changes may be made therein without departing from the broader spirit and scope of the invention as set forth in the appended claims.
For example, instead of the linear functions illustrated in
Also, the connections as discussed herein may be any type of connection suitable to transfer signals from or to the respective nodes, units or devices, for example via intermediate devices. Accordingly, unless implied or stated otherwise, the connections may for example be direct connections or indirect connections. The connections may be illustrated or described in reference to being a single connection, a plurality of connections, unidirectional connections, or bidirectional connections. However, different embodiments may vary the implementation of the connections. For example, separate unidirectional connections may be used rather than bidirectional connections and vice versa. Also, plurality of connections may be replaced with a single connections that transfers multiple signals serially or in a time multiplexed manner. Likewise, single connections carrying multiple signals may be separated out into various different connections carrying subsets of these signals. Therefore, many options exist for transferring signals.
Furthermore, each signal described herein may be designed as positive or negative logic. In the case of a negative logic signal, the signal is active low where the logically true state corresponds to a logic level zero. In the case of a positive logic signal, the signal is active high where the logically true state corresponds to a logic level one. Note that any of the signals described herein can be designed as either negative or positive logic signals. Therefore, in alternate embodiments, those signals described as positive logic signals may be implemented as negative logic signals, and those signals described as negative logic signals may be implemented as positive logic signals.
Furthermore, the terms “assert” or “set” and “negate” (or “deassert” or “clear”) are used herein when referring to the rendering of a signal, status bit, or similar apparatus into its logically true or logically false state, respectively. If the logically true state is a logic level one, the logically false state is a logic level zero. And if the logically true state is a logic level zero, the logically false state is a logic level one.
Those skilled in the art will further recognize that the boundaries between logic blocks are merely illustrative and that alternative embodiments may merge logic blocks or circuit elements or impose an alternate decomposition of functionality upon various logic blocks or circuit elements. Thus, it is to be understood that the architectures depicted herein are merely exemplary, and that in fact many other architectures can be implemented which achieve the same functionality. For example, the voice detectors and calculators shown in
Also for example, in one embodiment, the illustrated examples may be implemented as circuitry located on a single integrated circuit or within a same device. For example, the conference call system may be implemented as a suitable programmed processor, such as a general purpose microprocessor or a digital signal processor. Alternatively, the examples may be implemented as any number of separate integrated circuits or separate devices interconnected with each other in a suitable manner.
Also for example, the examples, or portions thereof, may implemented as soft or code representations of physical circuitry or of logical representations convertible into physical circuitry, such as in a hardware description language of any appropriate type.
However, other modifications, variations and alternatives are also possible. The specifications and drawings are, accordingly, to be regarded in an illustrative rather than in a restrictive sense.
In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word ‘comprising’ does not exclude the presence of other elements or steps then those listed in a claim. Furthermore, the terms “a” or “an,” as used herein, are defined as one or more than one. Also, the use of introductory phrases such as “at least one” and “one or more” in the claims should not be construed to imply that the introduction of another claim element by the indefinite articles “a” or “an” limits any particular claim containing such introduced claim element to inventions containing only one such element, even when the same claim includes the introductory phrases “one or more” or “at least one” and indefinite articles such as “a” or “an.” The same holds true for the use of definite articles. Unless stated otherwise, terms such as “first” and “second” are used to arbitrarily distinguish between the elements such terms describe. Thus, these terms are not necessarily intended to indicate temporal or other prioritization of such elements The mere fact that certain measures are recited in mutually different claims does not indicate that a combination of these measures cannot be used to advantage.
This application is a continuation of U.S. patent application Ser. No. 13/501,472, entitled “CONFERENCE CALL SYSTEM, METHOD, AND COMPUTER PROGRAM PRODUCT,” filed on Apr. 12, 2012, now U.S. Pat. No. 8,619,963, which is a National Stage Entry under 37 C.F.R. §371 of PCT/IB09/54943, filed Nov. 6, 2009, the disclosures of which are hereby expressly incorporated by reference in their entirety.
Number | Name | Date | Kind |
---|---|---|---|
4388717 | Burke | Jun 1983 | A |
4466094 | Renner | Aug 1984 | A |
4488291 | Eschmann et al. | Dec 1984 | A |
4495616 | Shuh | Jan 1985 | A |
4499578 | Marouf et al. | Feb 1985 | A |
4757493 | Yuen et al. | Jul 1988 | A |
4845499 | Bruce | Jul 1989 | A |
6369846 | Katsumi | Apr 2002 | B1 |
7330541 | Surazski et al. | Feb 2008 | B1 |
8436888 | Baldino et al. | May 2013 | B1 |
20020097857 | Kwan | Jul 2002 | A1 |
20020181686 | Howard et al. | Dec 2002 | A1 |
20030198195 | Li | Oct 2003 | A1 |
20050094795 | Rambo | May 2005 | A1 |
20060146735 | Shaffer et al. | Jul 2006 | A1 |
20070263821 | Shaffer et al. | Nov 2007 | A1 |
20080049921 | Davis et al. | Feb 2008 | A1 |
20090079815 | Baird | Mar 2009 | A1 |
20090220064 | Gorti et al. | Sep 2009 | A1 |
20090253418 | Makinen | Oct 2009 | A1 |
20100020954 | Gilg et al. | Jan 2010 | A1 |
20100165889 | Madabhushi et al. | Jul 2010 | A1 |
20110093273 | Lee et al. | Apr 2011 | A1 |
Number | Date | Country |
---|---|---|
2000-322753 | Nov 2000 | JP |
2008079024 | Apr 2008 | JP |
10-2000-0059487 | Oct 2000 | KR |
10-2007-0080017 | Aug 2007 | KR |
Entry |
---|
Prasad et al.; “On the Problem of Specifying the Number of Floors for a Voice-Only Conference on Packet Networks”; International Conference of Information Technology, Research and Education, 2003 (ITRE2003); Aug. 11-13, 2003; pp. 22-26;New Jersey, US. |
Prasad et al.; Automatic Addition and Deletion of Clients in VoIP Conferencing:; Sixth IEEE Symposium on Computers and Communications (ISCC); 2001; pp. 386-391. |
International Search Report mailed Aug. 6, 2010, for International Application No. PCT/IB20091054943, 3 pages. |
Number | Date | Country | |
---|---|---|---|
20140112464 A1 | Apr 2014 | US |
Number | Date | Country | |
---|---|---|---|
Parent | 13501472 | US | |
Child | 14143053 | US |