JPH04354246A - Telephone conference equipment - Google Patents

Abstract

PURPOSE:To configure a telephone conference with ease of use economically by utilizing asymmetry of an ATM line so as to connect plural telephone conference equipments with a single loop tying all points. CONSTITUTION:Each telephone conference equipment is accommodated in an ATM network by an ATM line 4-7 and connected to a loop with a line using an ATM cell. An in-network delay fluctuation of a received voice cell is absorbed by a delay fluctuation absorbing device 4-5 and sent to a voice signal replacement device 4-6 for each prescribed time. Voice signals of all equipments including its own equipment are in existence in the voice cell. The voice signal replacement device 4-6 writes a new voice signal stored in a transmission buffer 4-4 and sends it to an ATM network. On the other hand, a voice signal synthesizer 4-1 adds signals of the voice cell received from the delay fluctuation absorption device 405 except its own equipment for each voice sampling and reproduces it as the synthesized voice signal. Thus, each equipment listens to all equipments except its own equipment to realize the voice conference.

Description

[Detailed description of the invention]

0001

[Industrial Field of Application] The present invention is a teleconferencing device, that is, by using a plurality of telephones and a packet communication network, a local area network (LAN), or an ATM communication network, users of the above-mentioned plurality of telephones can communicate with each other. The present invention relates to a teleconference device that can be suitably used to implement a teleconference.

0002

[Prior Art] When a plurality of people (N people: members of a conference call) who are geographically distant from each other hold a conference call,
Additionally synthesizes the voices of m speakers (m = 1 in normal usage mode) and distributes the synthesized voices to N-m people,
A so-called audio bridge device is required. When implementing a conference call, how the audio bridge device is placed within the communication network or at the telephone terminal greatly influences the economic efficiency and convenience of the conference call service. As for the arrangement method of the audio bridge device, for example, NTT Research Practical Application Report Vol. 34 No. 4 "Voice Conference Communication System" (1985)
As disclosed in 2010), an intra-network voice bridging method and a terminal voice bridging method are known. The in-network voice bridge method is a method in which one or more voice bridge functions are installed in a network such as a switching network, separate from the terminal or telephone, and one voice bridge is usually shared by multiple users. . On the other hand, the terminal voice bridge method is a method in which a voice bridge function is installed in each telephone. The former intra-network voice bridging method requires the incorporation of a voice bridging function within the communication network, which requires adding or changing functions to the communication network, whereas the latter terminal voice bridging method requires this. Such additions and changes are not necessary and can be realized using only the functions of the terminal's audio addition and distribution circuit. Therefore, when using the terminal voice bridge method, it has excellent flexibility and mobility, and can easily implement teleconferences not only on public networks but also on private networks for companies. In the past, the mainstream method for realizing teleconferences was the intra-network bridging method, but in recent years, advances in digital signal processing LSIs such as DSPs have made it possible to realize voice bridging at low cost. As a result, it has become possible to put into practical use a terminal audio bridging method in which an audio bridging function is installed on each terminal.

0003

[Problem to be solved by the invention] Among the above-mentioned terminal audio bridge systems, the ones currently in practical use in Japan are:
There is a method that uses a meeting phone (hereinafter referred to as the "meeting phone method"). This method uses the ISDN basic interface 2B+D, and is used, for example, for "audio applications using ISDN"
TT Technology Journal, vol. 2, No. 10,199
Year 0). According to this, the meeting phone system has a three-way communication circuit consisting of two B channels (B1 channel, B2 channel), three audio adders as an audio bridge function, and an audio signal for each B channel. It has an encoder (COD) and a decoder (DEC) for transmitting and receiving (see FIG. 4). Furthermore, as a specific example, FIG. 5 shows the flow of transmitting and receiving audio signals when a four-party conference is held. In the configuration shown in FIG. 5, telephone B receives an audio signal (a) from telephone A, adds its own outgoing signal (b) with an audio adder, and transmits the signal to telephone C. On the other hand, receiving an audio signal (c+d) from telephone C,
It adds its own transmission signal (b) and transmits it to telephone A. When each terminal connected in series performs this operation, the transmission signal from telephone B is transmitted to all other telephones. Also, audio signals (a), (c+d) received from both channels
) to hear the outgoing signals of all other telephones. As described above, by using two channels for each telephone, it becomes possible to implement a conference call using a plurality of telephones. However, the conference phone system described above has the following drawbacks. That is, (1) each telephone requires two communication channels for both transmission and reception, and a normal telephone that does not have one communication channel cannot be used. This is also disadvantageous from the economic point of view of teleconferencing. (2) There is a large call delay. That is, in the case of the example shown in FIG. 5, telephone A must pass through the link (communication network) six times before it receives a response from telephone D to its own outgoing call. To provide a supplementary explanation of the latter, in the meeting phone system, when a conference is generally formed using N telephones, 2 (N-1) links are used. Call information in a conference call can be passed over multiple links using traditional analog links with low latency, or
Although this is not a practical problem for synchronous transfer mode (STM) links, it becomes a serious problem when using ATM links, where the duration is expected to be tens of milliseconds per link. For example, assuming that the delay per ATM link is 30 milliseconds and the number of conference call members N is 11, the response delay will be 600 milliseconds based on the above relationship. This delay 6
00 milliseconds is the upper limit for a natural conversation, and in this example, it would be inappropriate to have more than 11 members. When configuring a conference call using ATM links, it is clearly preferable to use a method that uses fewer links. The present invention was made in view of the above circumstances, and
The purpose is to solve the above-mentioned problems in the conventional technology, and to make it easy to use in the ATM communication network environment.
Another object of the present invention is to provide a teleconference device that makes it possible to economically organize a teleconference.

0004

[Means for Solving the Problems] The above objects of the present invention are as follows:
In a system that realizes a conference call by connecting multiple telephone conference devices in a loop through one communication channel, a specific field of the signal stored in multiple information fields in the received voice cell is The removed signal is converted into a linear PCM signal, and the PC
a means for adding and synthesizing the M signals, and a means for replacing the signal in the specific field in the received voice cell with a new voice signal of the same magnitude and transmitting the signal, one or temporally continuous. Storing the transmitted audio signal in an information field in a predetermined cell for each of the plurality of teleconference devices, so that the plurality of cells are dividedly used by the audio signals transmitted from the plurality of teleconference devices; A cell is sent to a transmission channel, and a plurality of audio signals corresponding to other teleconference devices are extracted from the cell received from a reception channel, and the received audio is synthesized and reproduced by adding these in digital format. This is accomplished by a teleconferencing device featuring the following features:

[0005]

[Operation] In the telephone conference device according to the present invention, ATM
Taking advantage of the asymmetry of the line, a configuration is adopted that connects all points with a single loop. Furthermore, a configuration is adopted in which audio signals from multiple predetermined points enter one audio cell (hereinafter, this is referred to as a "shared cell format"). Therefore, if there are many conference members, a plurality of temporally consecutive cells will be used. Teleconference device (“terminal”) according to the present invention
(or sometimes abbreviated as "equipment"), each terminal making up an audio conference must first determine where in the cell the audio signals sent by the terminal will be stored in order to start the conference call. It shall be determined in advance at the time of configuring the loop. That is, each terminal removes the voice signal of its own terminal stored one cycle ago from the cells (there may be multiple cells) received from the reception channel N-1.
Combine the audio signals of the devices into one (there may be multiple)
Play as audio. At the same time, the new voice signal generated by the own terminal is written to a predetermined position in the cell and transmitted. As a result, the members forming the audio conference can listen to the conversations of all the other members except their own voices, and the audio conference can be formed. As described above, in the telephone conference device according to the present invention, the number of links required in the meeting phone system described above is N, instead of 2 (N-1), and each device uses a total of 2 links. Since it is a book, it has the advantage that it is possible to make teleconferencing services more economical and that ordinary telephones can be used instead of special telephones. As mentioned above, the fact that a teleconference can be configured with a small number of links means that a teleconference can be easily implemented even in an ATM communication network with a large delay. As in the previous example, the delay per loop is 600 milliseconds, and the delay per ATM link is 30 milliseconds.
In terms of milliseconds, the maximum number of members in a teleconference that can be configured is 20, which is significantly increased from 11 members in the case of the meeting phone system.

[0006] Another feature of the teleconferencing device according to the present invention is that by using an audio signal in a format in which audio signals from a plurality of locations are carried in one cell, that is, the above-mentioned shared cell format, cell integration delay can be reduced. It is possible to reduce the For example, μ-lawPCM with 8kHz sampling,
The time for storing information in an ATM cell with an information field length of 48 octets (ie, cell formation time) is 6 milliseconds. This delay has a significant impact on the naturalness of telephone conversations. In contrast, the telephone conference device according to the present invention divides and uses one (or more) cells based on audio information from multiple locations as a shared cell format, so it is possible to shorten the cell integration time. It is. For example, when the number of shared users is 4, that is, when the cell is divided into four, the cell formation time is 1.5 milliseconds, and the above-mentioned influence on the naturalness of the conversation can be alleviated. In addition, in order to set the cell formation time to 1.5 milliseconds without adopting the carpooling format, only 1/4 of the cells will be applied to the audio signal, and the remaining part will be left vacant, so the cell formation time will be 1.5 milliseconds. This can be shortened, but this will lead to a deterioration in cell usage efficiency. Furthermore, in the telephone conference device according to the present invention, the audio signals from each member making up the audio conference are individually stored in one cell (or a plurality of consecutive cells) in a shared cell format. Therefore, each device can, as necessary, allow conversations with only a specific person or localize the other party's voice in multiple spatially different positions using multiple receivers (speakers) (so-called "sound field localization"). ), it is also possible to implement a function that facilitates speaker identification.

[0007]

Embodiments Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In the embodiment described below, a case will be explained in which μ-law PCM is used as audio information in a cell. Here, μ-lawPCM uses the statistical properties of voice amplitude to logarithmically transform and compress the amplitude and encode it, thereby obtaining high quality voice with a small amount of information. This is the encoding method used. Figure 2 shows an example of a telephone conference configured using an ATM network and four telephone conference devices.
This shows a situation in which four telephone conference devices A, B, C, and D are connected in a loop to conduct an audio conference. As will be described later, audio cells storing the audio signals of each teleconference device in each field circulate on the above-mentioned loop. Each teleconference device extracts the audio signal of its own teleconference device from the received voice cell, that is, the audio signal excluding the audio signal written one round before, and synthesizes and synthesizes it as received audio.
Reproduce. Thereafter, a new audio signal generated by the own telephone conference device is written in the field for the own telephone conference device of the cell, and is transmitted to the next telephone conference device.

FIG. 1 is a diagram showing the configuration of a telephone conference apparatus according to an embodiment of the present invention, and the flow of signals and processing in the configuration example shown in FIG. 2. In the figure, 4-1 is an audio signal that extracts all audio signals from the received audio cell except for the own teleconferencing device, converts it into a linear PCM signal, and adds and synthesizes the audio signals belonging to these multiple teleconference devices. A signal synthesizer, 4-2 a D/A converter that converts the synthesized linear PCM signal into audio, and 4-3 an A/D converter that converts the audio signal of the transmitter into the aforementioned μ-law PCM signal. , 4-4 is a transmission buffer that holds the audio signal to be stored in the transmission cell until the replacement audio cell arrives;
5 is a delay fluctuation absorption mechanism for temporarily accumulating voice cells received from the ATM network 4-7, absorbing delay fluctuations occurring within the network, and reproducing continuous voice; 4-6 is the delay fluctuation absorption mechanism; 4-7 is an audio signal replacer that replaces the own telephone conference device field of the audio cell output from 4-5 with the audio signal stored in the transmission buffer 4-4 and transmits it to the ATM network 4-7; An ATM line is shown that accommodates a receive channel for receiving ATM cells from a teleconferencing device and a transmitting channel for transmitting ATM cells to other teleconferencing devices.

The operation of this embodiment configured as described above will be explained below. Each teleconference device is accommodated in the ATM network by ATM lines 4-7, and is logically connected in a loop by logical lines using ATM cells. A
The intra-network delay fluctuations of voice cells received from the TM network are absorbed by the delay fluctuation absorbing mechanism 4-5, and sent to the voice signal replacer 4-6 at regular intervals. In this voice cell, the audio signals of all the teleconference devices including the own teleconference device exist. The audio signal replacer 4-6 writes and writes new audio signals accumulated in the transmission buffer 4-4.
Replace it and send it to the ATM network. On the other hand, the audio signal synthesizer 4
-1 converts the signals of the audio cell received from the delay fluctuation absorption mechanism 4-5 other than the fields of the own teleconferencing device into linear PCM signals, and performs arithmetic addition for each audio sampling in each field. Play as one (or more) synthesized audio signals. Therefore, each teleconference device can listen to the voices of all the teleconference devices other than its own, making it possible to implement an audio conference. In the above explanation, μ-lawPCM is used for voice information in a cell.
Although an example using ADPCM is shown, it is also possible to use other encoding methods such as ADPCM. However, in that case, the audio signal replacer 4-6 needs to function as an ADPCM encoder, and the audio signal synthesizer 4-1 needs to function as an ADPCM decoder.

[0010] Furthermore, in the above explanation, it is assumed that the audio signals at each point in the cell are stored separately in each field. It is also possible to localize and reproduce the sounds from the teleconference device at different spatial locations, thereby making it easier to identify the speaker. FIG. 3 is a diagram showing an example of the configuration of a cell used for the above-mentioned conference call. The time stamp in the conference call management information is used to compensate for delay fluctuations in the ATM network and to reproduce smooth audio. The number of conference members indicates the number of members making up the current audio conference, and this information determines how many cells should be transmitted at one transmission (or reception) trigger. The cell type is information that identifies whether the cell in question is the first cell, middle cell, or last cell, and if the number of conference members is small, there is no middle cell and it is the first cell. At the same time, it becomes the last cell. Regarding the order in which audio signals are stored in cells, for example, the host terminal that convened the audio conference assigns the audio signal to the field of the first cell (the #1 audio signal in Figure 3), and then assigns the audio signal in the order of loop connection. It's easy to understand what's going on.

As explained above, according to the above embodiment,
Utilizing the asymmetric nature of ATM lines, multiple teleconferencing devices are connected through a single loop that connects all points.
The number of links required to configure an audio conference can be reduced to N (N-1). As a result, the audio delay in an audio conference can be reduced from 2(N-1) Td to NTd. Note that here, Td indicates the delay per link. The above result shows that under the same allowable response delay conditions, an audio conference can be configured with 2 (1-1/N) times, that is, approximately twice the number of conference members. Also,
This means that if the number of conference members is the same, the configuration according to this embodiment can achieve a smaller response delay, that is, a more natural conversation. In addition, in the configuration according to this embodiment, the receiving side removes the own audio signal written in the previous cycle and synthesizes only the audio from other teleconferencing devices to create the reproduced audio. There is no audible echo, which does not cause deterioration of voice quality. Furthermore, as mentioned above, since the cell sharing system is used, the value of cell integration delay can be reduced.
Another effect is that it is possible to prevent an increase in audio delay that is harmful to conversation. It goes without saying that the above embodiment is an example of the present invention, and that the present invention is not limited thereto.

0012

[Effects of the Invention] As described in detail above, according to the present invention, a teleconference device is realized that is easy to use and enables economical construction of teleconferences in an ATM communication network environment. This has the remarkable effect that it can be done.

[0013]

[Brief explanation of the drawing]

FIG. 1 is a diagram showing the configuration of a teleconference device according to an embodiment of the present invention, and the flow of signals and processing in the configuration example shown in FIG. 2.

FIG. 2 is a diagram showing an example of configuring a teleconference using an ATM network and four teleconference devices according to the embodiment.

FIG. 3 is a diagram showing the configuration of an ATM cell used in the example.

FIG. 4 is a diagram showing the prior art, and is an explanatory diagram showing the configuration of a meeting phone type three-party communication circuit currently used.

5 is a diagram showing the state of communication when a terminal bridge type four-party conference is held using the meeting phone shown in FIG. 4; FIG.

[Explanation of symbols]

4-1: Audio signal synthesizer, 4-2: D/A converter, 4-
3: A/D converter, 4-4: Transmission buffer, 4-5: Delay fluctuation absorption mechanism, 4-6: Audio signal replacer, 4-7:
ATM line.

Claims (1)

[Claims] Claim 1: In a system in which a telephone conference is realized by connecting a plurality of telephone conference devices (telephones) in a loop through one communication channel, information stored in a plurality of divided information fields in a received voice cell is used. means for converting a signal excluding a specific field from the received signal into a linear PCM signal, and adding and synthesizing the PCM signal; means for replacing and transmitting an audio signal with an audio signal, such that one or a plurality of temporally consecutive cells are dividedly used by the audio signals sent from the multiple teleconference devices, storing a transmitted audio signal in an information field in a predetermined cell for each of the devices, transmitting the cell to a transmitting channel, and receiving a plurality of audio signals corresponding to other teleconferencing devices from said cell received from a receiving channel; extract,
A telephone conference device characterized by synthesizing and reproducing received voices by adding these in digital format.