JPS60136450A - Terminal equipment for packet switching - Google Patents

Abstract

PURPOSE:To improve efficiency of voice communication and economic efficiency by converting the aural signal produced from a speaking person to character information and transmitting it as a packet to a transmission line and synthesizing a voice in the reception side on a basis of character information converted to the packet. CONSTITUTION:After a speaking person A calls a controller CTL(B) connected to a packet switching network PSN by the operation of a key incorporated in a controller CTL(A) and is connected to the controller CTL(B), he inputs the voice to a telephone set T(a). The controller CTL(A) digitizes the aural signal and edits it to a packet signal of a prescribed format and transmits it to the packet switching network PSN, and it is transmitted to the controller CTL(B) of a called subscriber B. The controller CTL(B) reproduces the aural signal on a basis of this packet signal by D/A conversion and inputs it to a telephone set T(b). The packet signal is transmitted after being stored in a message communication system MCS. Thus, the quantity of information is reduced considerably to improve efficiency of communication and economic efficiency.

Description

DETAILED DESCRIPTION OF THE INVENTION (al) Technical Field of the Invention The present invention relates to a terminal device that performs voice communication via a packet-switched network, and particularly to a terminal device for packet switching that significantly reduces the redundancy included in voice and packetizes it. Regarding terminal devices.

(b) Technical background FIG. 1 is a diagram showing an example of a voice packet communication system to which the present invention is applied. In FIG. 1, the telephone v! A.T.
is accommodated in the packet switching network PSN via the control device CTL. The analog audio signal output from the phone fiT (A) when caller A sends a call is sent to the control device CTL.
(A) is converted into a digital signal, organized into packets in a predetermined format, and transmitted to the packet switching network PSH. The packet is transmitted via the packet switching network PSN to the control device CTL (B) corresponding to the interlocutor B, and after being converted again into an analog voice signal, the telephone T
It is transmitted to interlocutor B via (B). Also, control device C
The packets sent by TL (A) are sent to the packet switching network P.
After being stored in the message communication system MC3 accommodated in the SN, it is transmitted to the control device CTL (B). Since a digitized voice signal has a huge amount of information, for example 64 kilobits per second, reducing the amount of information in a digitized voice signal as much as possible will improve the efficiency and economy of voice packet communication. is necessary.

(C) Prior art and problems Conventional packet switching terminal equipment converts the analog voice signal output from the telephone T (A) by the caller into a digital signal, and then deletes the silent part. , the active portion was packetized and sent to the packet switching network PSN. Therefore, a voice signal with a high degree of redundancy is transmitted while the voice signal is being transmitted, which may reduce the efficiency and economy of voice communication by the packet switching terminal device.

Tdl OBJECTS OF THE INVENTION An object of the present invention is to eliminate the drawbacks of the conventional packet switching terminal devices as described above, and to improve the efficiency and economy of voice communication by the packet switching terminal devices.

(el) Structure of the Invention The object of the present invention is to provide a voice recognition means for recognizing a voice signal sent by a caller and converting it into text information, and a means for packetizing the text information output from the voice recognition means and transmitting the packet to a transmission line. This is achieved by providing a voice synthesis means that performs voice synthesis and outputs a voice signal based on the character information that arrives in packet form from the transmission path, and also recognizes the voice signal sent by the caller and outputs the character information. a speech recognition means for converting the text information into a packet; a means for packetizing the character information outputted from the speech recognition means and sending it to a transmission line; and □recognition result return means for inputting the character' information outputted by the voice recognition means to the voice synthesis means, so that the caller can receive the voice signal he/she sends. This is achieved by making it possible to check the recognition results, and further includes a voice recognition means that recognizes a voice signal sent by a caller and converts it into text information, and packetizes and transmits the text information output from the voice recognition means. This is achieved by providing means for transmitting the character information to the transmission line, and means for outputting the character information arriving in the form of packets from the transmission line in the form of characters.

(f) Examples of the invention An embodiment of the present invention will be described below with reference to the drawings.

FIG. 2 is a diagram showing a packet switching terminal device according to an embodiment of the present invention, and FIG. 3 is a diagram showing an example of signal waveforms at various parts in FIG. 2. Note that the same reference numerals indicate the same objects throughout the figures. In Figures 2 and 3,
The caller instructs the control device CTL to make a call by operating a call key (not shown) on the operation panel KP. Buffer BUF2 holds a call request transmitted from operation panel KP, and transmits an interrupt signal to central processing unit CPU.

The central processing unit CPU analyzes the received interrupt signal, recognizes that the interrupt request source is the buffer BUF2, and extracts the calling request held in the buffer BUF2. The central processing unit CPU that has analyzed the call request writes data for displaying dialing permission on the operation panel KP in the buffer BUP2. As a result, a dial enable indicator lamp (not shown) on the operation panel KP lights up. When the caller recognizes the lighting of the dialing indicator lamp and inputs the called party number from the dial keys (not shown) on the operation panel KP, the input dial information is stored in the memory MEM under the control of the central processing unit CPU. Once the dial is completed, the high-speed line control IC
The packet is organized into a call request packet in a predetermined format by U, and sent to the packet switching network PSN via the line handling unit LTU and line termination unit LSU. The packet switching network PSN that has received the call request packet confirms that the called packet switching terminal device is not in communication and transmits an incoming call indication packet. The control device CTL of the packet switching terminal device on the destination side, which has received the incoming call indication packet, displays the incoming call to the called party by sounding a buzzer (not shown) built in the operation panel KP. When the called party operates the call key on the operation panel KP to notify the calling party that the call has been accepted, the called party control unit CTL creates a call acknowledgment packet in the same process as described above, and performs packet exchange &M P Return it to SN. When the originating control unit CTL receives an incoming call acknowledgment packet from the packet switching network PSN via the line terminating unit LSU, the central processing unit CPU lights the communication ready lamp (not shown) on the operation panel KP through the same process as described above. At the same time, data for causing the speech recognition unit ARU to start recognition operation is written into the buffer BUFO. When the caller who recognizes the lighting of the communication ready lamp starts a call, the microphone M of the telephone T is activated.
The analog format audio signal v1 output by the audio recognition unit A
The signal is transmitted to the RU and the silence detector SDU.

The speech recognition unit ARU performs speech recognition processing on the analog-format speech signal 1 by a known method, outputs character information v2, and transmits it to the buffer BUFO. On the other hand, the silence detection unit SDU detects the sound state of the analog audio signal v1, and transmits the sound detection signal v3 to the buffer BUFO.

The buffer BUF0 holds the character information v2 received from the speech recognition unit AR1J and transmits an interrupt signal to the central processing unit CPU only when the voice detection signal v3 is received from the silence detection unit SDU. The central processing unit CPU that received the interrupt signal extracts the character information v2 from the buffer BUFO and stores it in the memory MEM. Further, the central processing unit CPU monitors the sound detection signal v3 outputted by the silence detection unit SDU at predetermined time intervals, and when a predetermined time to has elapsed since the sound detection signal v3 became silent, or when the sound detection signal v3 is output from the memory MEM. When the number of characters of the stored character information v2 reaches a predetermined number, the high-speed line control unit ICU is instructed to packetize the character information v2. The high-speed line control unit HCU extracts the character information v2 stored in the memory MEM, organizes it into packets in a predetermined format, and sends it to the line correspondence unit LTU and the line termination device L.
It is sent to the packet switching network PSN via the SU. In the destination control device CTL, the high-speed line control unit HCU extracts character information v2 from the packet that arrived from the packet switching network PSN via the line termination unit LSU under the control of the central processing unit CPU, and stores it in the memory MEM. do.

The central processing unit CPU extracts the character information v2 stored in the memories MP and M, and stores it in the buffer BUFI. The speech synthesis unit ASU converts the text information v2 stored in the buffer BUFI into an analog audio signal by a known method.
1 is synthesized and transmitted to the speaker S of the telephone T. As a result, the called party listens to the content of the caller's conversation. When the caller operates the call end key (not shown) on the operation panel KP to end the call, a call end packet and a call end confirmation packet are exchanged between the calling and receiving side control devices CTL, and the packet switching network PSN The communication path set for is released.

As is clear from the above description, according to this embodiment, the analog voice signal v1 from the caller's call is converted into text information V2 in the calling side control device CTL, and then packetized. Since the information is transmitted to the device CTL, the amount of information is significantly reduced. Since the called side control device CTL synthesizes an analog audio signal vl from the transmitted text information V2 and transmits it to the called party, the caller's tone of voice cannot be discerned, but the content of the call is transmitted to the called party. It becomes possible.

In the above-mentioned process, when the central processing unit CPU stores the character information v2 extracted from the buffer BUFO in the memory MEM and accumulates it in the buffer BUFI, the speech synthesis unit ASU converts the character information v2 into an analog-format audio signal through the above-mentioned process. v1 and sends it back to the caller via speaker S.

As a result, the calling party can confirm the recognition result of the content of his/her call. Furthermore, as mentioned above, the central processing unit CPU does not start packetizing the text information v2 based on the number of characters in the text information v2 stored in the memory MEM or the duration of the silent state of the voice detection signal v3, and instead converts the content of the call itself into packets. When it is detected that the caller operated the packetization instruction key (not shown) provided on the operation panel KP, the text information stored in the memory MEM is read for the first time. Packetization also allows the calling party to modify the recognition result of his/her call before sending it to the packet-switched network PSN.

In addition, the central processing unit CPU transmits the character information v2 extracted from the packet arriving from the packet switching network PSN or the character information v2 output from the voice recognition unit ARU to the display control unit CCU instead of transmitting it to the voice synthesis unit ASU.
By transmitting the character information v2 to the character information v2, it becomes possible to display the character information v2 in character form on the display device CRT, or to print it out using a printer device installed in conjunction with the character information v2.

Note that FIGS. 2 and 3 are only one embodiment of the present invention, and, for example, the configuration of the control device CTL is not limited to that shown in the figures, and a part of the control unit ficTL may be replaced with a general-purpose personal computer. Although many other modifications are possible, such as configuring the system using a computer, the effects of the present invention remain the same in either case.

(Illusion) As described above, according to the present invention, it is possible to significantly improve the efficiency and economy of voice communication using a packet switching terminal device.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing an example of a voice packet communication system to which the present invention is applied, FIG. 2 is a diagram showing a packet switching terminal device according to an embodiment of the present invention, and FIG. It is a figure which shows an example of the signal waveform of each part. In the figure, ARU is a speech recognition unit, ASU is a speech synthesis unit, BUFO to BUF2 are buffers, CCU is a display control unit, CPU is a central processing unit, CRT is a display device, CTL is a control unit, FCU is a floppy disk control unit, FD is a floppy disk device, ICU is a high-speed line control unit, KP is an operation panel, LSU is a line termination unit, LTU is a line support unit, M is a microphone, MC3 is a message communication system, MEM is a memory, PCTL is a packet control unit, PSN is a packet switching network, 1S is a speaker, SDU is a silence detector, T is a telephone, and
is a predetermined time, vl is an analog audio signal, v2 is text information, and v3 is a voice detection signal. 3 2

Claims (1)

[Scope of Claims] (11) Voice g! Orihira means for converting the voice signal sent by the caller into textual information by tH6, and the textual information outputted by the voice recognition means is packetized and transmitted to the transmission path. 7) A terminal device for exchange, characterized in that it is provided with a voice synthesizing means for performing voice synthesis and outputting a voice signal based on the character information arriving in the form of a packet from the transmission path. (2) a voice recognition means for recognizing the voice signal sent by the caller and converting it into text information; a means for packetizing the text information output from the voice recognition means and transmitting the packet to a transmission path;
a speech synthesis means for performing speech synthesis and outputting a speech signal based on the text information arriving in packet form from the transmission path; and a recognition result return means for inputting the text information output from the speech recognition means to the speech synthesis means. 1. A packet switching terminal device comprising: a terminal device for packet switching, wherein the caller is able to check the recognition result of the voice signal transmitted by the caller. (3) a voice recognition means for recognizing a voice signal sent by a caller and converting it into text information; a means for packetizing the text information output from the voice recognition means and transmitting the packet to a transmission line; A terminal device for packet exchange, characterized in that it is provided with means for outputting the character information arriving in a packet format in a character format.