JPH06303305A - Voice responding telephone set - Google Patents

Abstract

(57) [Summary] [Object] An object of the present invention is to improve the recognition rate of a response word in a voice response telephone as much as possible, with regard to a voice response telephone equipped with a voice call means and a voice response means. [Configuration] When a user executes a predetermined test operation, the telephone (100) is set to a pseudo incoming call state (eg, a speaker (101) outputs a ringing sound) and a pseudo incoming call setting means (110) , Pseudo incoming call setting means (11
0) sets the telephone (100) in the pseudo incoming call state, reproduces the response word registered in the voice response means (104) and outputs it from the speaker (101) and the response word reproduction means (120) and the response. The voice response means (1) outputs the voice uttered by the user who listens to the response word reproduced by the word reproduction means (120).
04) and [a recognition result is repeated a predetermined number of times if unsuccessful] and a response result notifying means (130) for notifying the user of the recognition result is provided.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a voice response telephone having a voice call means and a voice response means.

As one of the diversified telephones, a loudspeaker type telephone which allows a user to talk by using a speaker and a microphone without a handset has been put into practical use under the name of hands-free telephone or the like. .

[0003]

2. Description of the Related Art In this type of loudspeaker type telephone, a response word such as "high" or "moshimoshi" is registered in advance, and when a user receives a call, the user inputs the response word from a microphone. A voice response unit that sets a response state without operating a response button such as a voice response button provided on a public address type telephone or performing a response operation such as raising a handset to start a call is, for example, the invention "voice recognition". Automatic answering telephone "(Japanese Patent Laid-Open No. 61-202550, published on September 8, 1986) and the like.

[0004]

However, even if the user utters the same response word, it is not always possible to utter in the same tone as the registered response word, and therefore the recognition rate of the response word is not always the same. The cost was not high, and there was a risk of inconvenience such as inability to respond even if the response word was uttered many times.

It has been experimentally confirmed that the user can always speak in a manner close to the registered response words by training the response words so that the recognition rate of the response words is improved. However, it is difficult for the user to voluntarily train the vocalization of a response word because it is uncertain when to call the telephone.

An object of the present invention is to improve the recognition rate of a response word in a voice response telephone as much as possible.

[0007]

FIG. 1 is a diagram showing the principle of the present invention. In FIG. 1, 100 is a telephone which is the subject of the present invention, 101 is a speaker, 102 is a microphone,
Reference numeral 103 is a voice communication means, 104 is a voice response means, and 200
Is a telephone line to which the telephone set (100) is connected.

110 is a telephone (100) according to the present invention.
It is a pseudo incoming call setting means provided in. 120 is a response word reproducing means provided in the telephone set (100) according to the present invention.

Reference numeral 130 indicates a telephone (100) according to the present invention.
Is a response result notifying means provided in the.

[0010]

The pseudo incoming call setting means (110) is provided with a telephone (10) when the user executes a predetermined test operation.
0) is set to the pseudo incoming state.

The response word reproducing means (120) reproduces the response word registered in the voice response means (104) in a state where the pseudo incoming call setting means (110) sets the telephone (100) in the pseudo incoming call state, It is sent from the speaker (101).

The response result notifying means (130) recognizes the voice uttered by the user who hears the response word reproduced by the response word reproducing means (120) by the voice response means (104), and the recognition result is recognized by the user. To notify.

It is considered that the pseudo incoming call setting means (110) causes the speaker (101) to output a pseudo call sound when the telephone (100) is set to the pseudo incoming call state.

The response result notifying means (130) repeats recognition of the response word issued by the user a predetermined number of times when the user's utterance cannot be recognized as the response word,
When all the recognitions are unsuccessful, it is considered to notify the user of the unsuccessful recognition of the response word.

Further, it is considered that the response result notifying means (130) notifies the recognition result of the utterance of the user by a predetermined sound indicating success or failure. Therefore, the user can at any time perform the vocal training of the response word registered in the telephone and can confirm the training result, so that the voice recognition rate of the voice response telephone is improved.

[0016]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. 2 is a diagram showing a voice response telephone according to an embodiment of the present invention, and FIG. 3 is a diagram showing an example of a voice recognition test process in FIG.

In FIG. 2, the telephone set (1
00) is a voice response telephone, and telephone line (1) is shown as telephone line (200) in FIG.
A speaker (4) is shown as the speaker (101) in FIG. 1, and the microphone (10) in FIG.
A microphone (5) is shown as 2), and a voice call circuit (3) is used as the voice call means (103) in FIG.
1 and a voice memory (6), a recording unit (7) and a voice recognition unit (9) are shown as the voice response unit (104) in FIG. 1, and the pseudo incoming call setting unit (110) in FIG. As a voice recognition test unit (11), a pseudo incoming call setting unit (12) and a test key (13) are provided,
Further, a reproducing section (8) is provided as the response word reproducing means (120) in FIG. 1, the voice memory (6) and the voice communication circuit (3) are used, and the voice is used as the response result notifying means (130) in FIG. The recognition test unit (11) and the pseudo incoming call setting unit (12) are used.

In the voice memory (6), the response word (A) previously inputted by the user from the microphone (5) [in the previous example, "high" or "moshimoshi", etc.) is recorded in the voice communication circuit (3) and recorded. Registered via department (7).

In this state, when an incoming call is received from the telephone line (1), the telephone circuit (2) detects a ringing signal arriving from the telephone line (1) and notifies the control unit (10) of the detection of an incoming call, the control unit (10) notifies the voice communication circuit (3) of the incoming call, causes the speaker (4) to acoustically output a calling signal, and notifies the user of the incoming call.

When the user who is aware of the incoming call utters a predetermined response word (A) to the microphone (5), the voice input from the microphone (5) passes through the loudspeaker communication circuit (3). Is transmitted to the voice recognition unit (9).

The voice recognition unit (9) is a voice communication circuit (3).
By collating the voice transmitted from the voice with the response word (A) extracted from the voice memory (6), it is recognized whether the utterance of the user is the response word (A) and the recognition result is controlled. Notify the department (10).

When the recognition result notified from the voice recognition unit (9) is successful recognition of the response word (A), the control unit (10) connects the telephone circuit (2) to the telephone line (1). On the other hand, after setting the answering state, and stopping the sound output of the ringing signal from the loudspeaker communication circuit (3) to the speaker (4), the loudspeaker communication circuit (3) is expanded by the speaker (4) and the microphone (5). Set to a state where you can talk.

When the user performs vocal training for the response word (A), he or she operates the test key (13). Control unit (1
0) activates the voice recognition test section (11) when detecting the operation of the test key (13).

The activated voice recognition test unit (11) activates the voice communication circuit (3) and the reproduction unit (8) and extracts the response word (A) from the voice memory (6) [step S in FIG. 3].
1], input to the reproduction unit (8) to reproduce the response word (A),
The loudspeaker (4) outputs the signal through the loudspeaker circuit (3) [step S2].

When the response word (A) has been reproduced and output, the voice recognition test unit (11) activates the pseudo incoming call setting unit (12) to the telephone line (1) to the voice communication circuit (3). From the speaker (4) is acoustically output [step S3], and this is repeated at intervals of 2 seconds until voice is received from the microphone (5) [step S4 and S5].

After listening to the calling sound, the user tries to utter a response word (A) having the same tone as that of the response word (A) previously output from the speaker (4). In the voice response telephone, when the voice uttered by the user is input from the microphone (5) [step S4], the voice input by the voice recognition unit (9) in the same process as described above is the response word (A). The recognition process is executed [step S6] and the recognition result is notified to the control unit (10).

In the control unit (10), the voice recognition test unit (11) analyzes the recognition result notified from the voice recognition unit (9) [step S7], and when the response word (A) is successfully recognized. , The pseudo incoming call setting section (12) is caused to generate a recognition success notification sound and output from the speaker (4) through the loudspeaker communication circuit (3) [step S8].

If the recognition of the response word (A) is unsuccessful, the voice recognition test section (11) continues until the predetermined number of repetitions (N) is reached (step S9), the pseudo incoming call setting section ( A pseudo call tone is input to the voice communication circuit (3) from 12) and is output from the speaker (4) to cause the user to repeat the utterance of the response word (A).

Each time the user listens to the calling sound,
The response word (A) previously output from the speaker (4) was tried and the response word (A) previously output from the speaker (4) was tried while training was attempted. The response word (A) having the same tone as that can be uttered, and the recognition is successful.

If the user does not succeed in recognizing the response word (A) even after repeating the utterance a predetermined number of times (N) [step S9], the voice recognition test section (11) makes a pseudo test. A sound for recognition failure notification is generated in the incoming call setting section (12) and is output from the speaker (4) via the loudspeaker communication circuit (3) [step S10].

As is clear from the above description, according to the present embodiment, the user operates the test key (13) provided on the voice response telephone to input the response word (A) which is registered in advance. After listening, listening to pseudo-ring tone and response word (A)
It becomes possible to perform training with the utterance of, and it becomes possible to utter the same tone as the registered response word (A) during the training, and the recognition rate of the response word (A) can be improved.

2 and 3 are merely embodiments of the present invention, and the configuration of the telephone (100) is not limited to that shown in the drawings, and the speaker (4) and the microphone ( Many other modifications can be considered such as 5) and the like which are separated from the others and connected by wireless communication means, but the effects of the present invention are not changed in any case. In the case of a mobile phone, a remote key can be provided by providing a test key (13) on the child device.

[0033]

As described above, according to the present invention, in the above-mentioned telephone, the user can perform the utterance training of the response word registered in the telephone at any time, and the training result can be confirmed.
The voice recognition rate of the voice response telephone is improved.

[Brief description of drawings]

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

FIG. 2 is a diagram showing a voice response telephone according to an embodiment of the present invention.

FIG. 3 is a diagram showing an example of a voice recognition test process in FIG.

[Explanation of symbols]

 1, 200 Telephone line 2 Telephone circuit 3 Loudspeaker communication circuit 4, 101 Speaker 5, 102 Microphone 6 Voice memory 7 Recording unit 8 Playback unit 9 Voice recognition unit 10 Control unit 11 Voice recognition test unit 12 Pseudo incoming call setting unit 13 Test key 100 Telephone set 103 Voice call unit 104 Voice response unit 110 Pseudo incoming call setting unit 120 Response word reproduction unit 130 Response result notification unit

Claims (4)

[Claims] 1. A voice call means (103) for allowing a user to talk using a speaker (101) and a microphone (102) without having a handset, and a telephone line (20).
0), when the user receives a call from the microphone (10
A voice response means (10) for recognizing a voice input from 2) with a previously registered response word as a reference and responding to the incoming call.
And a pseudo incoming call setting means (110) for setting the telephone (100) to a pseudo incoming call state when the user executes a predetermined test operation. The pseudo incoming call setting means (110) causes the telephone (10
0) is set to the pseudo incoming call state, the response word registered in the voice response means (104) is reproduced, and the response word reproduction means (120) is transmitted from the speaker (101).
And a response result notification in which the voice response unit (104) recognizes a voice uttered by the user who listens to the response word reproduced by the response word reproduction unit (120) and notifies the recognition result to the user. And a means (130) for providing a voice response telephone. 2. The pseudo incoming call setting means (110), when the telephone (100) is set to a pseudo incoming call state,
The voice response telephone according to claim 1, wherein a pseudo ringing sound is transmitted from the speaker (101). 3. The response result notifying means (130) repeats recognition of a response word issued by the user a predetermined number of times when the utterance of the user cannot be recognized as the response word. The voice response telephone according to claim 1, wherein if all of the recognitions are unsuccessful, the user is notified of the unsuccessful recognition of the response word. 4. The voice according to claim 1, wherein the response result notifying means (130) notifies the recognition result of the utterance of the user by a predetermined sound indicating success or failure. Answering phone.