JPH04287558A - Voice communication system for physically handicapped person - Google Patents

Abstract

PURPOSE:To allow parties of speech impediment or having difficulty in daily conversation to come to understanding with each other in voice freely through the use of a voice communication equipment such as a subscriber telephone set or a public set with respect to the voice communication system for an impediment via a telephone terminal equipment. CONSTITUTION:A telephone terminal equipment is provided with an IC card reader 1, a voice analysis section 2 and a voice correction section 3. An IC card in which a voice pattern or a voiceprint of the user having difficulty of usual conversation such as speech impediment is registered is read from the IC card reader and the voice inputted to the voice input section by the user is collated with the content read from the IC card at the voice analysis section 2 to confirm that the user corresponds to the user registered in the card. All voice signals generated by the user are recognized by a voice correction section 3 and the signal is converted into a usual voice corresponding to the result of recognition to make conversation.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a voice communication system for persons with disabilities such as speech disabilities. In recent years, the development of communication means and methods has been remarkable in the development of personal computer communication, facsimiles, and telephones. At the same time, new storage media such as IC cards are being developed and becoming popular. On the other hand,
Communication methods based on text input have been considered and are being implemented in some areas for people with disabilities and an aging society, but there is a need for the development of natural-speech communication methods that can be used more widely. There is.

0002

[Prior Art] In the conventional voice communication system, I
C cards are used in the following fields: That is, in order to retrieve notes taken during a conversation or recorded message information while the person is away, ID verification, starting a telephone call using a specific operating procedure, or accumulation of billing data are performed. these are,
The main users are ordinary people. However, for people with language disabilities or those who have difficulty in normal conversation, there are situations in which they cannot use IC cards as a means of communication because they have a disability in language as a means of communication. Therefore, communication methods other than language, such as personal computer communication using text, communication by facsimile, or communication means other than language, in which text input is converted into speech and transmitted, are increasingly being used.

0003

[Problem to be Solved by the Invention] As mentioned above, when a person with a language disorder or who has difficulty in normal conversation wants to obtain a means of communication by voice, it is necessary to either transmit the real voice without direct voice conversion or to input text. A voice conversion method is used, but there are problems such as it takes time to select words that are easy to pronounce using the human voice, and the pronunciation is difficult for the other party to hear, making it difficult to convey information smoothly. . In addition, character input has problems such as limited expressions, time required for input, and limited devices. The present invention provides a voice communication system for people with disabilities that allows people with language disabilities or disabilities in normal conversation to communicate freely through voice communication using voice communication devices such as subscriber telephones and public telephones. With the goal.

0004

[Means for Solving the Problems] FIG. 1 is a diagram showing the basic configuration of the present invention. In FIG. 1, 1 is an IC card reader that reads the IC card 6 and transfers it to the voice analysis section, 2 is a voice analysis section that receives and analyzes the input voice and the content transferred from the IC card reader, and 3 is an analysis a voice correction unit that converts the recognized voice into normal voice; 4 is a voice input unit; 5 is a voice output unit; 6 is an IC card carried by the user and in which the user's voice pattern or voiceprint is registered in advance;
7 is an alarm output unit that outputs a buzzer or the like when the voice analysis results cannot be analyzed.

The present invention adds an IC card reader and a voice analysis/conversion device to a commonly used voice communication device such as a telephone, and when an IC card with voice patterns and voiceprints registered is input, voice analysis is performed. It is a device that conveys one's intentions to the other party by converting it into normal voice and outputting it.

[0006]

[Operation] When the voice communication device shown in FIG.
The registered voice pattern or voiceprint is read and transferred to the voice analysis section 2. When a user inputs voice from the voice input unit 4 (inputting unclear voice as is), the voice analysis unit 2 checks whether the characteristics of the user's voice match by comparing the input voice with the voice pattern or voiceprint. , if they match, the subsequent input speech of the caller is supplied to the speech correction section 3.

[0007] In the speech correction section 3, first the speech recognition device 30
By comparing the voice with the stored words and expression formats, the input voice is recognized using data for analysis taken out from the IC card 6, and the recognition result is supplied to the voice conversion device 31 to determine whether it is normal. The audio is converted into audio and sent out from the audio sending unit 5. In the speech analysis section 2,
If a discrepancy is detected, the alarm output unit 7 is activated to output a buzzer, etc. to notify that the call will become a normal call, and at the same time, the switch (SW) on the output side of the voice analysis unit 2 is switched to change the input voice. The audio signal is supplied to the audio output section 5 as it is. It also informs that the user is not the real user of the IC card 6.

[0008]

Embodiment FIG. 2 is a configuration diagram of an embodiment, and FIGS. 3 to 5 are operational flows (1) to (3) in the configuration of the embodiment. Figure 2
1 to 3 and 6 are the same as in FIG. 1, 1 is an IC card reader, 2 is a voice analysis section, 3 is a voice correction section, and 6 is an I.
C card, 7 is a buzzer sending control unit (corresponding to the alarm output unit in Fig. 1), 8 is a handset consisting of a transmitter 8a (corresponding to the audio input unit 4 in Fig. 1) and a receiver 8b, and 9 is a telephone receiver. A/D converter, 10 is an amplifier, 11 is a telephone line (L01, L02
12 is a call detecting unit (including the voice transmitting unit 5 in Fig. 1) that performs the operation of detecting outgoing and incoming calls for the line (indicated by ), and sends voice and dial signals and receives voice from the line. A bell activation control section operated by a signal, 13 represents a bell that is rung by the bell activation control section.

[0009] The speech analysis unit 2 includes an I
An internal memory (not shown) receives voice pattern or voiceprint data representing the user's characteristics from the data read out when the IC card 6 is inserted by the C card reader 1.
and outputs it to the verification circuit 2b. On the other hand, when the user's voice (predetermined voice, for example, "moshi, mo") is uttered from the transmitter 8a of the handset 8, the voice input is sent to the reception unit 2a via the A/D converter 9. The voice pattern or voiceprint is extracted and supplied to the verification circuit 2b. The matching circuit 2b matches (compares) the voice pattern or voiceprint input from the IC card reader with the voice pattern or voiceprint uttered by the user, and analyzes whether there is a certain correlation between the two.

The voice correction section 3 is composed of a voice recognition device 30 and a voice conversion device 31 as in FIG. Voice recognition device 3
0 can be configured using conventional speech recognition technology,
In the example of FIG. 2, the judgment unit 30c compares the matching data 30a stored in the IC card from the IC card reader 1 with the digital data of the voice input from the receiver 8b, thereby recognizing what kind of voice the voice is. . This recognition result is input to the speech conversion device 31 and converted into normal speech.

In this example, the speech conversion device 31 includes a speech dictionary 31a and a D/A converter 31b, and a speech analysis section 2.
When you receive the result recognized in
) or audio data corresponding to the word (PCM code)
The corresponding voice data is taken out from the voice dictionary 31a storing the voice data, converted into an analog signal by the D/A converter 31b, and output from the call detection unit 11 to the telephone line via the amplifier 10.

Operation flows (1) to (3) in the configuration of FIG. 2 are shown in FIGS. 3 to 5, and will be outlined below. When you first take the handset 8 (see Figure 2) off-hook,
The IC card reader 1 starts up and becomes ready for use, and the IC
When the card is inserted into the IC card reader 1, the registered voice pattern or voiceprint is read out from the card (40 to 43 in FIG. 3). Next, the read contents are transferred to the reception section 2a of the speech analysis section 2, and a waiting state is established. Thereafter, the user performs normal connection processing such as inputting the dial of the other party (44 to 46).

[0013] When the connection process is performed, a connection with the other party is established by the operation of a telephone exchange (not shown). Moving on to Figure 4,
When the other party answers (47 in Figure 4), the calling party (the user of this device) speaks, and the voice is input to the voice reception unit 2a (48, 49 in Figure 4), the voice pattern of the IC card. and the input voice are sent to the matching circuit 2b (see 50), and the two are compared to see if they match (there is a correlation) (see 51).
), and if there is a mismatch, activate the buzzer sending control unit (52).
), a buzzer is sent to the caller through the handset 8b, and the call is switched to a normal call (53). This switching is performed by driving the switch SW shown in FIG. Sent via .

When a match is detected in the above matching, the speech pattern is transferred to the speech recognition device 30 (54), the words, expression format, etc. of the input speech are identified (55), and the process moves to FIG. When the identification result is notified to the voice conversion device 31 (56), it is converted into a corresponding normal voice by the voice dictionary 31a in FIG. It is sent to the telephone line (57).

[0015] Step 47 (FIG. 5) to Step 5
The operation No. 7 is repeatedly executed while the conversation is in progress. When the handset 8 is on-hook and the call is ended (same 5
8), the IC card reader has finished operating (59), and the IC
The card is ejected from the insertion slot (60), and the operation of this voice communication device is completed.

Although the above-described embodiment is constructed by providing an additional device to a normal general telephone or public telephone, the same function can be realized by providing an additional device to a cordless telephone. In that case, it is appropriate to have an IC card reader, voice analysis section, and voice conversion section provided in the base unit of the cordless phone, but they may also be provided in the handset side. Furthermore, if a similar configuration is added to a radio device (Handy Talkie, etc.) that performs voice communication between specific persons, it will become possible for persons with speech disabilities to use the device.

[0017]

[Effects of the Invention] According to the present invention, people with language disabilities or people who have difficulty in normal conversation can register the voice characteristics or voiceprints of each individual on an IC card in advance, and use the voice analysis/conversion device to create a voice. It becomes possible to communicate with ordinary people by sending out natural voices using communication devices.

[0018]

[Brief explanation of the drawing]

FIG. 1 is a diagram showing the principle configuration of the present invention.

FIG. 2 is a configuration diagram of an embodiment.

FIG. 3 is an operation flow (1) in the configuration of the embodiment.

FIG. 4 is an operation flow (2) in the configuration of the embodiment.

FIG. 5 is an operation flow (3) in the configuration of the embodiment.

[Explanation of symbols]

1 IC card reader 2　　　　　　　　Speech analysis section 3 Audio correction section 4 Audio input section 5 Audio transmission section 6 IC card 7 Alarm output section

Claims (2)

[Claims] [Claim 1] In a voice communication system for people with disabilities via a telephone terminal, the telephone terminal is equipped with an IC card reader, a voice analysis section, and a voice correction section, and the voice pattern of a user who has difficulty in normal conversation, such as a language disorder, is recorded. Or an IC with registered voiceprint
The card is read by an IC card reader, and a voice analysis unit compares the voice input by the user from the voice input unit with the content read from the IC card to confirm that the user is the one corresponding to the card. A voice communication system for persons with disabilities characterized in that all generated voices are recognized in a voice correction unit and converted into normal voices according to the recognition results for conversation. [Claim 2] In claim 1, when the speech analysis section confirms that the user is compatible with the IC card, the speech correction section converts the pre-registered specific expression format and words of the input speech into normal forms. A voice communication system for people with disabilities that is characterized by converting it into voice for conversation.