JPH06232979A - Multi-media communication conference system - Google Patents

Abstract

PURPOSE:To attach a token only with an easy operation and to make a conference progress smoothly in a multi-media communication conference system. CONSTITUTION:Voices (words) in accordance with multi-media terminals 32-33 are registered in advance on a main station terminal 28. The fact that a call originating request is issued from the multi-media terminal 32 is notified to a multi-medial terminal 28 by touching a token request area on the LCD screen of the multi-media terminal 32. The multi-media terminal 28 displays from which participating terminal the request is issued on an LCD 29 screen, and a name equivalent to a terminal on which the token is desired to attach and registered in advance is vocalized after a token attaching area is touched. Thereby, the name is converted to data representing the feature of the voice by the feature extraction part of a speech recognition part, and furthermore, it is sent to a similarity calculation part, and data with highest similarity out of dictionary data is selected, then, the token is attached on a recognition result.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a multimedia communication conferencing system, and more particularly to a multimedia communication conferencing system for assigning tokens among a plurality of multimedia communication terminals which are multipoint connected by communication lines.

[0002]

2. Description of the Related Art A multimedia device is a communication terminal device capable of compositely using a plurality of media,
It is possible to input / output still image information, moving image information, voice information, handwriting information (telewriting information), and the like. By connecting a plurality of the multimedia devices via a communication line such as ISDN, it is possible to construct a multimedia communication conference system capable of holding a communication conference between remote persons.

However, if a plurality of multimedia devices arbitrarily input still image information, moving image information, audio information or drawing information, information is confused.
Tokens are added to the communication conference in an orderly manner, and only the multimedia device having the token has the right to speak. The operation of token addition in a multimedia device is generally manually input using an operation panel.

Further, for example, in Japanese Patent Laid-Open No. 2-265346, the right to speak is automatically given to the request for the right to speak from the conference participating stations.

[0005]

However, in such a conventional multimedia communication conference system, when the token is manually input using the operation panel, the number of communication terminals participating in the conference becomes large. There is a problem in that it takes time and effort to find a target terminal from the operation panel, resulting in poor operability.

[0006] Further, there is a problem in that the one which automatically assigns the token is impossible even if the speaker is desired to be designated by the chairperson terminal or the like. The present invention has been made in view of the above-mentioned conventional problems, and when assigning a token to a participating station of a communication conference terminal, it is possible to assign the token by a simple operation, and communication is performed smoothly. An object of the present invention is to provide a multimedia communication conference system capable of proceeding a conference.

[0007]

The invention according to claim 1 is
It is configured by a plurality of multimedia communication terminals capable of compositely using a plurality of media, and one terminal of the plurality of multimedia communication terminals is a master station terminal,
When the remaining terminal is used as a participating station terminal and a token request is made from the participating station terminal to the main station terminal and a token is added from the main station terminal to the participating station terminal, the participating station terminal to which the token is added In a multimedia communication conference system in which information to be transmitted is displayed on each communication terminal, the main station terminal has a voice recognition means for recognizing a token request from a participating station terminal by voice, and the voice of the voice recognition means. The feature is that tokens are added based on the recognition result.

According to a second aspect of the present invention, in the multimedia communication conference system according to the first aspect, the voice recognition means of the main station terminal extracts the voice feature of the token request and converts it into data showing the feature. The feature extraction unit and the data indicating the feature of the voice are input from the feature extraction unit, the similarity between the input data and the data in the dictionary data stored in advance is calculated, and the dictionary having the highest similarity is input. It is characterized in that it comprises a similarity calculation unit that selects data in the data and outputs it as a voice recognition result.

The invention according to claim 3 is the invention according to claim 1 or 2.
In the multimedia communication conference system described above, the main station terminal has a voice registration means for registering a voice corresponding to each participating station terminal, and each participating station terminal is controlled based on the voice registered in the voice registration means. It is characterized by identification.

[0010]

In the invention according to claim 1, when the main station terminal gives a token to the participating station terminal, the participating station terminal to which the token is given can be designated by the voice recognition means. It is easy to operate, and the conference can proceed smoothly.

According to the second aspect of the present invention, since the voice recognition means is composed of the feature extraction unit and the similarity calculation unit, the voice feature is extracted in determining the participating station terminal to which the token is added, It is accurately determined based on the result of calculating the similarity. In the invention described in claim 3, since the main station terminal is provided with the voice registration means, the voice corresponding to each participating station terminal is registered in advance, and each participating station terminal is controlled based on the registered voice. Accurately identified.

[0012]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below with reference to the drawings. First, the configuration will be described. FIG. 1 is a block diagram showing the configuration of a multimedia terminal of a multimedia communication conference system according to an embodiment of the present invention.

In FIG. 1, a CPU (Central Processi)
The ng unit) 1 controls the entire system of the multimedia device according to a program written in a ROM (Read Only Memory) 2. RAM (Random Acc
The ess Memory) 3 stores a work area and data necessary for the operation of the CPU 1.

A DMA (Direct Memory Access) controller 4 controls the transfer operation of image information in the DMA transfer mode. The handset 5 is a handset that exchanges voice data. Audio codec (CODEC)
6 decodes a digital voice signal received through the ISDN line and converts it into an analog voice signal, or encodes the analog voice signal and transmits it as a digital voice signal.

The microphone 7 is a voice input device for transmitting conversation contents during the conference. The voice recognition unit 8 is for recognizing voice information input from a microphone. The printer interface 10 connects the printer 9 and the system bus 27, and prints and prints based on the input print data.

The scanner interface 12 connects the scanner 11 to the system bus 27 to connect the scanner 11 to the system bus 27.
The document data read in is sent to the system bus 27. The document memory 13 stores document data and the like in a writable / readable manner. The timer 14 manages the current time. Interrupt (INT) controller 23
Performs interrupt control of the CPU 1.

The communication control unit 16 executes transmission of image data including telewriting data through an NCU (network control unit) 17 having an S interface with an ISDN line according to a predetermined transmission control procedure. It should be noted that this communication control unit 16 uses at least the B channel (64 Kbi
t / sec) It is configured so that simultaneous communication using two lines is possible.

The codec (CODEC) 18 encodes image information to be transmitted by a predetermined method and compresses the amount of the information, and also decodes the image information encoded at the time of reception to restore the original image information. . Graphic controller 1
Reference numeral 9 controls the graphic data accumulated in the video RAM (VIDEO RAM) 20.

Liquid crystal display / touch panel (LCD / T
P) The controller 21 includes the LCD 22 and the touch panel 2
3 is connected to provide a user interface or L
With respect to the document data displayed on the CD 22, the terminal having the token controls the drawing data and the like when the drawing information is input using the touch panel 23 and the touch pen.
The touch panel 23 is placed on the surface of an LCD (liquid crystal display device), and the operator touches the screen of the LCD to input handwritten information or LC in this system.
The command displayed on the D screen can be selected.

The moving picture codec 24 performs coding / decoding of video information, and is connected with a video (VIDEO) camera 25 and a television monitor 26. The system bus 27 is a signal line for exchanging data by connecting the above units. The multimedia terminal of this embodiment is
It is composed of these blocks and has a still image transmission function, a telewriting function, a telephone function, a data communication function and the like.

FIG. 2 is a diagram showing a configuration example of a multimedia communication conference system in which a plurality of the multimedia terminals of FIG. 1 are connected via a communication network, and FIG. 3 is an example of an LCD screen of a participating station terminal during a conference. FIG. 4 is a diagram showing an example of the LCD screen of the main station terminal during the conference. As shown in FIG. 2, the conference participants can use the LCD 2 of the multimedia terminal 28.
A conference can be held while looking at the screen of 9 and a conference participant with a token can write handwritten information on the LCD screen.

Here, the operation of this embodiment when the multimedia terminal 28 is the main station terminal and the multimedia terminals 32 to nth multimedia terminal 33 are the participating station terminals will be described with reference to FIGS. This will be described with reference to the flowchart of FIG. 5 is a flowchart showing an operation procedure at the time of voice registration, and FIG. 6 is a flowchart showing an operation procedure at the time of voice recognition.

First, the master station terminal 28 registers voices (words) corresponding to the multimedia terminals 32 to 33. The specific operation method is, for example, when registering the multimedia terminal 32 in the planning department as the registration partner, as shown in FIG. 5, “Registration” displayed on the LCD 29 screen of the main station terminal 28. A menu (not shown) is touched (step 100) to select the "multimedia terminal 3" which is the other party's terminal.
2 "is designated (step 101). This information is LC
It is sent to the CPU 1 via the D / TP controller 21.

Next, touch "Start" (step 102) and say a word corresponding to the other party's terminal, for example, "planning department" (step 103). This audio signal is
It is sent from the microphone 7 to the voice recognition unit 8, and the feature extraction unit extracts the features (step 104) to create a dictionary pattern (step 105). This data is stored in the memory as a dictionary in association with the registered terminal (step 106). Similarly, words corresponding to n participating station terminals are registered. These series of registration controls are
This is done by the CPU 1. This registration work is completed before the communication conference.

Next, the operation of giving a token during a conference will be described. The multimedia terminals 28 and 32-33 of FIG.
Up to n terminals are connected via a communication network,
The same screen is displayed to the conference participants on the screen of the CD (see FIGS. 3 and 4). This screen is a document at the time of the meeting or a memo during the meeting. Further, on the screen of the participating station terminals of FIG. 3, a screen for command input to the multimedia terminal such as a token request from the participating station terminal to the main station terminal is also displayed at the same time.

Here, in the case where the operator of the participating station terminal of the multimedia terminal 32 makes a token request to the multimedia terminal 28 which is the main station terminal, the LC shown in FIG.
Touch the token request area 35 on the D22 screen.
As a result, the multimedia terminal 28, which is the main station terminal, is informed that the multimedia terminal 32, which is a participating station terminal, has made a speech request. The multimedia terminal 28 displays from which participating terminal terminal the request came from on the screen of the LCD 22 as shown in FIG. Of course, if there is a token request from another terminal,
Display the data of the terminal that made the request. In the multimedia terminal 28, which is the master station terminal, of the terminals for which a token request has been made, the token registration area 37 is touched with a pre-registered name corresponding to the terminal to which the token is to be given. I will speak later. This voice signal is sent to the voice recognition unit 8 via the microphone 7 and converted into data indicating the voice feature in the feature extraction unit. Further, it is sent to the similarity calculation unit, the dictionary data having the highest similarity is selected, and it is the recognition result. Then, this recognition result is sent to the CPU 1, and a token is added to this recognition result.

The feature extraction unit of the present embodiment includes a feature extraction LS
I is used, and a recognition processing LSI is used for the similarity calculation unit. The feature extraction LSI extracts a power spectrum from the input voice, extracts a peak on the frequency from the power spectrum, performs binarization processing of "0" and "1" based on the extracted peak, and outputs a BTSP (Binary Time). Sp
ectrum Pattern) Generates data.

The recognition processing LSI is composed of a CMOS standard cell, and BTSP data and 15-channel power spectrum data are input by interrupt processing by a periodic pulse from the feature extraction LSI. In the voice section detection, time = frequency pattern (TSP:
Time Spectrum Pattern) data is used in the interrupt process according to the periodic pulse. Then, the BTPS data is input according to the voice section signal generated there.

As described above, according to the multimedia communication conference system of the present embodiment, when the main station terminal gives tokens to the participating station terminals, the token giving function can be designated by voice without manual input. , Easy to operate, now you can smoothly proceed with the meeting.

[0030]

According to the invention described in claim 1, when the main station terminal gives the token to the participating station terminals, the participant station terminal to which the token is given can be designated by the voice recognition means, so that the token manually input The operation is easier than the giving operation, and the conference can be smoothly progressed.

According to the second aspect of the invention, since the voice recognition means is composed of the feature extraction unit and the similarity calculation unit, the voice feature is extracted when determining the participating station terminal to which the token is added. However, it is possible to make an accurate determination based on the result of calculating the degree of similarity. According to the invention described in claim 3, since the main station terminal is provided with the voice registration means, the voice corresponding to each participating station terminal is registered, and each participating station is based on this registered voice. The terminal can be accurately identified.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a multimedia terminal of a multimedia communication conference system according to an embodiment of the present invention.

FIG. 2 is a diagram showing a configuration example of a multimedia communication conference system in which a plurality of multimedia terminals of FIG. 1 are connected via a communication network.

FIG. 3 is a diagram showing an example of an LCD screen of a participating station terminal during a conference.

FIG. 4 is a diagram showing an example of an LCD screen of a main station terminal during a conference.

FIG. 5 is a flowchart showing an operation procedure at the time of voice registration.

FIG. 6 is a flowchart showing an operation procedure at the time of voice recognition.

[Explanation of symbols]

 1 CPU 2 ROM 3 RAM 8 voice recognition unit 16 communication control unit 17 NCU 19 graphic controller 20 video RAM 21 LCD / TP controller 22 LCD 23 touch panel 28, 32, 33 multimedia terminal 31 communication line network

Claims (3)

[Claims] 1. A plurality of multimedia communication terminals capable of composite use of a plurality of media, wherein one terminal of the plurality of multimedia communication terminals is a main station terminal and the remaining terminals are As a participating station terminal, when a token request is made from the participating station terminal to the main station terminal and a token is given from the main station terminal to the participating station terminal, information sent from the participating station terminal to which the token is given Is displayed on each communication terminal, the main station terminal has voice recognition means for recognizing a token request from a participating station terminal by voice, and based on the voice recognition result of the voice recognition means. A multimedia communication conferencing system characterized in that a token is added to the multimedia communication conference system. 2. The multimedia communication conference system according to claim 1, wherein the voice recognition means of the master station terminal extracts a voice feature of the token request, and converts it into data indicating the feature. Data indicating a voice feature is input from the feature extraction unit, the similarity between the input data and the data in the dictionary data stored in advance is calculated, and the data in the dictionary data with the highest similarity is selected. Then, a multimedia communication conference system, comprising: a similarity calculation unit that outputs the result as a voice recognition result. 3. The multimedia communication conference system according to claim 1, wherein the main station terminal has a voice registration means for registering a voice corresponding to each participating station terminal, and is registered in the voice registration means. A multimedia communication conferencing system characterized in that each participating station terminal is identified based on the audio.