JP3322140B2 - Voice guidance device for vehicles - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a voice guidance system for a vehicle which outputs a voice into a vehicle and outputs a voice again in response to a request for re-outputting the voice.

[0002]

2. Description of the Related Art Conventionally, a navigation device for performing route guidance is known, and vehicles equipped with the navigation device are increasing. In this navigation device, by setting a destination, an optimal route from the current position to the destination is searched for and stored as a travel route. Then, while detecting the current position, the current position and the traveling route are displayed on a map including the current position to provide route guidance. At a guidance intersection where the vehicle turns left or right, guidance is provided by enlarging and displaying the traveling direction at the intersection. Such a navigation device facilitates selection of a traveling direction in traveling to a destination.

[0003] In addition, optical beacons and radio beacons are installed on the roadside, and traffic congestion information and the like on the road in the traveling direction are flown from there, and a system for providing congestion information and the like by FM multiplex broadcasting has been put into practical use. I have. If such information is used, more suitable route guidance can be performed by the navigation device.

[0004] Guidance on turning right and left by the navigation device is performed during traveling. The driver gazes ahead while driving, and guidance such as turning right and left is preferably provided by voice. Therefore, a device that outputs voice guidance such as "Across this road to the left in the left direction."

[0005] Here, voice guidance is greatly affected by the situation of surrounding noise and the like. That is, if the noise is loud, the probability that the voice guidance cannot be heard increases. Therefore,
In the apparatus described in Japanese Patent Application Laid-Open No. 3-209600, it is estimated from the response of the user (usually a driver) to the voice guidance and the noise situation at that time whether or not the voice guidance could not be heard. Then, the voice guidance is output again. As a result, when it is estimated that the user could not hear the voice, the voice guidance is automatically performed again, and more reliable guidance is achieved.

[0006]

However, in the above conventional example, whether or not to output sound again is determined by uniform judgment. Therefore, it is not possible to respond to individual differences in listening ability. In other words, for people with low hearing ability, lower the criteria for not being able to hear, and for people with high listening ability, raise the criteria for not being able to hear, and perform the guidance output again for each individual user. Could not.

SUMMARY OF THE INVENTION The present invention has been made to solve the above problems, and an object of the present invention is to provide a vehicular voice guidance device capable of performing re-voice output according to a user's listening ability. And

[0008]

SUMMARY OF THE INVENTION The present invention relates to a voice guidance apparatus for a vehicle, which outputs a voice into a vehicle and outputs a voice again in response to a request to output the voice again. A re-sound output means for identifying the reason for the output request and outputting a re-sound corresponding to the reason; and a re-sound output validity for evaluating the re-sound output validity according to the presence or absence of a re-re-output request after the re-sound output. And a voice output control means for storing a relationship between the re-voice and the reason when it is determined to be valid and reflecting the relationship in a subsequent voice output.
The control means describes the corresponding form for each of the multiple reasons.
Remember, in the subsequent audio output,
Call the corresponding form stored according to the
It is characterized by outputting a voice again in a state .

As described above, when re-outputting is performed, the validity of re-voice output is determined based on whether or not there is a request for re-outputting.
For example, if there is no re-re-output request, it is determined that re-output is appropriate. Then, the relation between the valid re-voice output and the re-output request is stored. Therefore, when there is a re-output request for the same reason, appropriate re-sound output can be performed based on the stored relationship. Therefore, it is possible to reduce the possibility that re-re-output is required and perform suitable re-voice output. In particular,
Since the relationship between the re-sound and the reason is stored based on the actual operation result, the re-sound output according to the user's listening ability is possible.

[0010] In particular, the audio output control means includes a plurality of logics.
The corresponding form corresponding to each of the reasons is stored, and the subsequent audio output
Remembers when you force, according to the reason you specified at that time
Call the corresponding form and output the sound again in the called form
You. In addition, the reason for the specified re-output request includes a hand
Preferably it includes that the dollar is in operation
You.

For example, if the noise inside the vehicle is large, it is preferable to increase the volume of the re-output sound, and if an esoteric expression such as a difficult place name is included, the expression should be changed to a plain expression. If the user is operating the steering wheel or the like, the user may simply output the sound again. By preparing a corresponding form corresponding to such a reason and performing re-voice output,
Appropriate re-sound output can be performed.

Further, the present invention is characterized in that, when a plurality of reasons are specified, re-sound output is performed in a plurality of forms corresponding to the plurality of reasons.

For example, when a loud and difficult expression is included, the sound volume can be increased and the re-sound output can be performed in a plain expression. Therefore, the re-sound output can be made appropriate.

Further, the present invention is characterized in that when the frequency of the re-output request under the same condition is a predetermined value or more, the audio output corresponding to the above-mentioned reason is executed from the first audio output.

If there are repeated re-output requests, it is preferable to output the sound in the form of re-sound output from the beginning. This can reduce the frequency of re-output request utterances,
Audio output more in accordance with the user's request can be performed.

[0016]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention (hereinafter referred to as embodiments) will be described below with reference to the drawings. FIG. 1 is a block diagram showing the overall configuration of the device according to the present embodiment. The navigation system 10 is connected to a current position acquisition device 12 such as a GPS and a map data storage device 14, and performs various processes for route guidance. In other words, according to the setting of the destination, the optimum route is searched, the route and the current position are displayed on a map on the display, and the guidance is provided at an intersection (guidance intersection) where the vehicle turns right or left while the route is set. Display an enlarged map for navigation and provide route guidance.

The navigation system 10 includes:
A speaker 20 is connected via a voice synthesis control unit 16 and a voice output control unit 18. And at the guidance intersection,
A guidance voice for turning right and left is output from the speaker 20. For example, 300 m before the guidance intersection, the voice synthesis control unit 16 and the voice output control unit 18 synthesize a voice such as “Guide to the right at 300 m ahead XX intersection” and output it from the speaker 20.

Then, the speech synthesis control unit 16
The guidance voice control device 22 is connected, and the estimation / guidance voice control device 22 changes the voice guidance to be output to an appropriate one according to the estimation result such as the reason why the re-input of the voice guidance is requested. To this end, a comparison analysis device 24 is connected to the estimation / guidance voice control device 22. The comparison analysis device 24 determines various countermeasures according to the reason for re-outputting the voice guidance, and based on the determination result, The operation of the estimation / guidance voice control device 22 is controlled accordingly.

A status and output request storage device 26 is connected to the comparison analysis device 24. The status and output request storage device 26 store the status of a request for re-outputting voice guidance in the past. . Therefore, the comparison analysis device 24 can determine various countermeasures with reference to the past situation. When a new situation occurs, the comparison analysis device 2
4 stores the status and the like in the status and output request storage device 26. The comparison analysis device 24 is connected to various parameter operation units 28 connected to the estimation / guidance voice control device 22.・
Various parameters in the guidance voice control device 22 are operated. Thereby, the volume of the audio output and the like are controlled.

An output request analyzer 32 having a built-in counter 30 is connected to the comparison analyzer 24. A voice input device 34 and a re-voice button 36 are connected to the output request analysis device 32. The voice input device recognizes a voice guide re-output request based on a user's utterance. The re-voice button 36 recognizes a voice guide re-output request based on a user operation. Then, the output request analysis device 32 recognizes the request for re-outputting the guidance voice based on the signals from the voice input device 34 and the re-voice button 36. The counter 30 counts the number of re-output requests.

The output request analyzing device 32 is connected to a status judging device 38.
Analyzes the reason for the re-output request according to the status from the status determination device 38. An audio volume level sensor 40, a noise sensor 42, a steering wheel operation sensor 44, an audio operation sensor 46, and a microphone 48 are connected to the situation judging device 38. Recognize environmental conditions such as whether the device is operating, operating the audio, or using the microphone, and determine the situation.

Here, the processing in the comparison analysis device 24 will be described with reference to the flowchart of FIG. First, it is determined whether or not a repetition input has been made (S11). This is performed by judging from a signal from the output request analyzing device 32 whether or not a re-output request by voice via the voice input device 34 or a re-output request by operating the re-voice button 36 has been made. Then, when there is no input of the reflection, there is no problem in the voice output, so that preparation for the next utterance is made.

On the other hand, if there is a repetition input (re-output request), it is compared with the status and the status stored in the output request storage device 26 (this status corresponds to the reason for the re-output request) ( S12). That is, the situation and the output request storage device 26 store the correct countermeasures that have been taken so far corresponding to the situation at that time. Therefore, if there is a coincident situation, the countermeasures may be the same. Therefore, if the situation and the situation that matches the output request storage device 26 are not stored, the reason is estimated by the reason estimation process (S13).
To skip.

Then, in S12, NO or S
When the reason is estimated and the re-output method (measure) is determined in step 13, a measure re-sound is generated based on the determined measure (S14).

When the re-sound is output in this way, it is determined whether a re-output request has been issued next (S15). That is, if the re-sound output is correctly performed, there is no request for re-output. Therefore, whether or not the re-sound output is correct (validity) is determined based on the presence or absence of the re-sound output request for the re-output.

In S15, if YES, that is, if there is a re-re-output request, it means that re-output did not meet the user's request, and various parameters for voice output are changed (S16). That is, various parameters for the audio output are changed, and the state of the audio output is determined again.

On the other hand, if there is no re-output request in S15, it is determined whether such a situation has occurred for the first time (S17). If this is the first situation, the relationship between the output method and the situation is stored in the situation and output request storage device 26 (S18). That is, the fact that there is no re-re-output request means that re-output was appropriate, and it is desirable to take the same countermeasures when a similar situation (reason) occurs. Therefore, this is memorized. Therefore, in S12, an appropriate countermeasure can be detected, and the probability that a re-output request is made can be reduced. In particular, since the relationship between the situation and the countermeasure is stored based on the response of the user, the countermeasure is determined based on the stored contents, so that a countermeasure according to the characteristics of the user is possible.

On the other hand, if it is not the first time in S17, the count value of the counter 30 is incremented by +
1 is performed (S19). Then, it is determined whether the count value obtained by adding 1 is equal to or more than a predetermined count value (a predetermined number of times, for example, three times) (S20). If this count value is high, the user is always issuing a re-output request in the same situation,
It is better to change the output method. Therefore, in the case of YES in S20, the first output method is changed (S20).
21). That is, the output method is changed so that a coping method considered to be correct in the situation where the re-re-output request is issued can be performed from the first time. This can reduce the probability that a re-output request is made. Note that the count may be reset when there is a request for re-output in the same situation.

In this way, the feedback (re-output request)
If there is, the contents are determined, re-sound output is performed, and if there is a re-re-output request, various parameters for re-sound output are changed, and if the re-re-output request is repeated, , Change the first output method. As a result, it is possible to perform voice guidance output more suited to the user's preference.

Next, the reason estimation process in S13 will be described with reference to FIG. First, the reason for the re-output request to be estimated is as follows: (a) I could not hear because the background sound was loud. (B) I could not understand because it was difficult to express. (C) I was talking or operating the device. It is possible that they missed it because they did. Then, measures for re-sound output should be different depending on such a reason.

That is, if the reason (a) is used, the sound volume is increased, if (b) is used, the expression is changed, and if (c) is used, the output is simply re-outputted in S14.

Therefore, these reasons are estimated in the reason estimation process of S13. First, the noise / audio level is determined (S31). That is, when the noise in the vehicle or the volume of the audio is at a level equal to or higher than a certain level, it is highly likely that the sound was not heard. Therefore, by this determination, the contribution of the reason (a) can be estimated. Therefore, according to the result determined by the situation determination device 38 according to the outputs of the audio volume level sensor 40 and the noise sensor 42, the levels of these levels are detected. Then, table comparison is performed based on the detected level (S32), and the reason (a) is compared.
Is given (S33). As shown in FIG. 4, a table is prepared for the inaudibility h (estimated evaluation amount of the reason (a)) corresponding to the audio volume and the noise amount, and the reason (a) is determined based on comparison with this table. ) Is determined.

When the estimated evaluation amount for the background sound volume is determined in this way, the guidance text is determined next (S34). That is, when the guide text is difficult, it is highly likely that the guide text could not be understood.
In particular, the more specific the guide text, the more difficult it is to express,
It will be difficult to understand. In addition, it is highly likely that a guide including a unique word (for example, a place name or a foreign language) could not be understood. Therefore, a table is prepared for the intelligibility of the guide text as shown in FIG. 5, the output guide text is compared with this table (S35), and the estimated evaluation amount of the reason (b) is given (S36). .

Next, a conversation / voicing judgment is made (S3).
7). That is, during conversation, attention is reduced, and there is a high possibility that the guidance output will be missed. Therefore, from the output of the microphone 48, the situation of conversation / speaking in the vehicle is taken in, and it is determined whether the user is talking or speaking. Then, the state of the occurrence is indicated by the amount of conversation (for example, the average volume within a predetermined time picked up by a microphone).
Is compared with the conversation level and the estimated evaluation value table as shown in FIG. 6 (S38), and the estimated evaluation amount for the reason (c) is determined (S39).

Next, the operation of the steering wheel, audio and the like is determined (S40). It is considered that the attention to the guidance output is reduced even when some operation is performed. Therefore, the operation state is detected according to the outputs of the steering wheel operation sensor 44 and the audio operation sensor 46. And FIG.
A table of the magnitude of the operation amount and the estimated evaluation amount as described above is prepared, and the estimated evaluation amount for the reason (c) is determined by comparing the table with the table (S41) (S42). here,
Since both S39 and S42 are the estimated evaluation amounts for the reason (c), the estimated evaluation amounts for the reason (c) may be the sum of the two.

As described above, since the estimated evaluation amount for each reason is obtained, a comparative calculation of these is performed (S4).
3) Estimate the reason (S44).

Although the final reason may be one, it is preferable to deal with a plurality of reasons. For example, as a countermeasure against the reason (a), the estimated evaluation amount h may be used as a parameter for determining the volume when re-outputting. That is, as shown in FIG. 4, in the region of the estimated evaluation amount h0 in which the audio volume and the noise amount are equal to or less than the predetermined value, the region of the estimated evaluation amount h1 in which the coefficient k3 and the audio volume and the noise amount are slightly larger than the predetermined value. Now, the coefficient k2,
The coefficient k1 is set in a region of the estimated evaluation amount h2 where the audio volume and the noise amount are larger than the above range, and the coefficient k0 is set in a region of the estimated evaluation amount h3 where the audio volume and the noise amount are further larger than the above range. Then, based on the set coefficient k, the re-output volume increase is determined by ΔV = k × V0 (where k is a coefficient, ΔV is the re-output volume increase, and V0 is the output volume). Also,
When the estimated evaluation amount for the reason (b) is large, the guide text is changed to a simpler one. Further, when the estimated evaluation amount for the reason (c) is large, the output of the same guidance voice is simply repeated. The countermeasure in S14 is performed by controlling the output of the various parameter operation units 28 via the estimation / guidance voice control device 22.

[0038] The reason for not being able to hear is that the speed is too fast or words that are difficult to hear. In such a case, it is also preferable to take measures such as judging this from the contents of the guidance text and slowing down the reading speed or changing to a guidance text output with a clear tongue.

Here, a re-output request switch as shown in FIG. 8 may be provided in place of the re-voice button 36 so that the user can input a countermeasure. That is, buttons such as "again", "louder", "slowly", and "paraphrase" are provided. If "again", the guidance output is simply repeated, and if "louder", the volume is increased by one level. If it is "slowly", the expression of the guidance is made slower, and if it is "paraphrased", it is changed to another expression (plain expression). With this method, appropriate measures can be taken. In this case, S1 in FIG.
In steps 4 and 15, a measure may be determined based on the operation of the operation button.

Further, as a method of changing the expression of the guidance, there are a method of using a plain text, a method of using a redundant expression, and a method of using a dialect, a buzzword, or a conventional expression. Furthermore, it is also preferable that the user can select an expression method.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating an overall configuration of an embodiment.

FIG. 2 is a flowchart showing an operation of a re-utterance process.

FIG. 3 is a flowchart illustrating an operation of a reason estimation process.

FIG. 4 is a diagram showing a relationship between a noise amount / audio volume and an estimated evaluation amount.

FIG. 5 is a diagram showing a relationship between the content of a guide text and an estimated evaluation amount.

FIG. 6 is a diagram illustrating a relationship between a conversation level and an estimated evaluation amount.

FIG. 7 is a diagram showing a relationship between an audio handle operation amount and an estimated evaluation amount.

FIG. 8 is a diagram illustrating a configuration example of a re-voice request switch.

[Explanation of symbols]

Reference Signs List 10 navigation system, 12 current position acquisition device, 14 map data storage device, 16 voice synthesis control unit, 18 voice output control unit, 20 speaker, 22 estimation / guidance voice control device, 24 comparative analysis device, 26 storage of status and output request Device, 28 various parameter operation units, 30 counters, 32 output request analyzers, 34
Voice input device, 36 re-voice button, 38 status determination device, 40 audio volume level sensor, 42 noise sensor, 44 handle operation sensor, 46 audio operation sensor, 48 microphone, 50 re-output request switch.

──────────────────────────────────────────────────続 き Continued on the front page (58) Field surveyed (Int.Cl. ⁷ , DB name) G10L 13/00

Claims (4)

(57) [Claims] An audio guidance device for a vehicle that outputs a voice into a vehicle and outputs a re-voice in response to a request for re-output of the voice, and specifies a reason for the re-output request in accordance with the voice output. Re-speech output means for outputting re-speech corresponding to this reason, re-speech output validity evaluation means for evaluating the re-speech output validity according to the presence or absence of a re-speech output request after re-speech output, Voice output control means for storing a relationship between the re-voice and the reason in the case of being performed, and reflecting the relationship in the subsequent voice output , wherein the voice output control means responds to each of a plurality of reasons.
The corresponding correspondence form is memorized, and at the time of the subsequent audio output,
Recalls the stored corresponding form according to the specified reason
And outputting a voice again in a called form . The apparatus according to the claim 1, wherein the reason for the re-output request to be identified, the handle operation
A voice guidance device for a vehicle, wherein the voice guidance device includes a message that the vehicle is inside . 3. The vehicular audio system according to claim 1, wherein when a plurality of reasons are specified, a re-sound output is performed in a plurality of corresponding forms according to the plurality of reasons. Guide device. 4. A voice output to a vehicle, and the voice is output.
Vehicle sound plan that re-outputs sound in response to re-output request
The internal device identifies the reason for the re-output request according to the audio output, and
Re-sound output means for outputting a re-sound corresponding to, and re-sound output according to the presence / absence of a re-send request after the re-sound output
Re-speech output validity evaluation means to evaluate the validity of the voice and the relationship between the re-voice and the reason when it is determined to be valid are stored
And has a sound output control means for reflecting the audio output after the, if the frequency of the re-output request in the same conditions is equal to or larger than a predetermined value, executes a sound output corresponding to the reason the first voice output A voice guidance device for a vehicle, comprising: