JP2017005369A - On-vehicle hands-free speech level estimation device, navigation device, and on-vehicle hands-free device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain an on-vehicle hands-free speech level estimation device for enabling a user to recognize the deterioration level of voice in hands-free speech.SOLUTION: A vehicle condition acquisition section 31 acquires a vehicle condition being information related to a travel speed, a road surface state, a window opening/closing state, an air-conditioner operation state, and speech reception voice output volume set in a speaker 12, etc. A hands-free performance estimation section 32 estimates the deterioration level of voice in hands-free speech with the use of voice processed by a hands-free voice processing section 15 and the vehicle condition. A notification section 33 notifies a user of the deterioration level of the voice estimated by the hands-free performance estimation section 32.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an apparatus for notifying a user of the state of a hands-free call used in a vehicle.

In a vehicle, there are many opportunities for voice to be processed, such as voice recognition for a user to interact with an in-vehicle device and hands-free call for a user to make a phone call without hindrance to driving.
During a hands-free call in the vehicle, not only the voice of the user who is the speaker, but also the running speed, road surface condition, window open / closed state, air conditioner operating condition, the voice of the listener output from the speaker, etc. Noise is input to the microphone. The sound input to the microphone is appropriately subjected to voice processing or the like for suppressing the influence of noise. However, in some cases, it is unavoidable that the sound quality reaching the listener is deteriorated. However, conventionally, the user has not been able to know what kind of sound quality finally reaches the receiver. In other words, even if it was difficult to hear the voice, the user could not notice it unless the listener pointed out.
By the way, when the user interacts with the in-vehicle device using voice recognition, the noise input to the microphone affects the voice recognition rate. In order to inform the user of the influence of noise, for example, Patent Document 1 discloses a technology for recording background noise before a user's utterance, predicting the performance level of voice recognition based on the recorded background noise, and notifying the user. Have been described.
Since hands-free calls are also processed in the same way as voice recognition, for example, it is conceivable to estimate the sound quality heard by the listener during a hands-free call using the technique of Patent Document 1, for example.

Special table 2008-501991 gazette

In speech recognition, it is important to increase the recognition rate, and it is necessary to notify the user whether speech that can be recognized on the device side is input. On the other hand, in a hands-free call, it is important to transmit a sound that is easy to hear for hearing, and it is necessary to notify the user how much sound is heard by the other party for hearing. As described above, the processing to be evaluated is different between the voice recognition and the hands-free call. For example, just because the voice has a high recognition rate, it does not necessarily mean that the voice is easy to hear. Therefore, even if the technique of Patent Document 1 is applied to a hands-free call, the sound quality heard by the receiver cannot be properly notified to the user.
In a hands-free call, the received voice that is the voice of the receiver output from the speaker may be input to the microphone. However, since such a received voice does not occur in voice recognition, Patent Document 1 does not consider the influence of the received voice. In this respect as well, it is not appropriate to apply the technique of Patent Document 1 to hands-free calling.

  The present invention has been made to solve the above-described problems, and an object thereof is to provide an in-vehicle hands-free call level estimation device that allows the user to know the deterioration level of voice during a hands-free call. To do.

  The in-vehicle hands-free call level estimation device according to the present invention includes a vehicle condition acquisition unit that acquires a vehicle condition indicating a state of an element that degrades a voice heard by a listener during a hands-free call, and a voice emitted by a user. A hands-free performance estimating unit for estimating a deterioration level of a voice heard by a listener during a hands-free call using at least a sound subjected to voice processing and a vehicle condition acquired by a vehicle condition acquiring unit; And a notification unit that outputs the degradation level estimated by the free performance estimation unit.

  According to the present invention, the user can know the voice degradation level during a hands-free call.

It is a figure which shows the structure of the vehicle-mounted hands-free call level estimation apparatus which concerns on Embodiment 1 of this invention. It is a figure which shows the hardware structural example of the vehicle-mounted hands-free call level estimation apparatus which concerns on Embodiment 1 of this invention. It is a flowchart which shows the process of the vehicle-mounted hands-free call level estimation apparatus which concerns on Embodiment 1 of this invention.

Embodiment 1 FIG.
FIG. 1 shows the configuration of an in-vehicle hands-free call level estimation device according to Embodiment 1 of the present invention. FIG. 1 also shows the in-vehicle hands-free device 1 and the voice communication device 2.
The in-vehicle hands-free device 1 includes a microphone 11 that converts input sound into an electric signal and outputs the sound, a speaker 12 that outputs the input electric signal as a sound wave, and a microphone 11 and the speaker 12 that are connected to the electric signal. A control unit 13 for processing and a voice communication interface unit 14 for connecting to the voice communication device 2 are provided. The control unit 13 includes a hands-free sound processing unit 15 that performs sound processing on the electric signal output from the microphone 11 and the electric signal output to the speaker 12, and various types used by the hands-free sound processing unit 15 for sound processing. And a parameter storage unit 16 for storing parameters.

The voice communicator 2 transmits the user's voice collected by the microphone 11 and subjected to voice processing by the control unit 13 to the other party who is a receiver via a communication network (not shown). In many cases, a sound including a sound other than the user's voice such as noise is transmitted. In addition, the voice uttered by the receiver is received via a communication network (not shown) and sent to the in-vehicle hands-free device 1. The voice of the listener is then output from the speaker 12.
The voice communication device 2 is, for example, a mobile phone, and is connected to the in-vehicle hands-free device 1, for example, Bluetooth (registered trademark: description is omitted below).

The in-vehicle hands-free call level estimation device 3 includes a vehicle condition acquisition unit 31, a hands-free performance estimation unit 32, a notification unit 33, and a deterioration level improvement unit 34.
The vehicle condition acquisition unit 31 acquires vehicle conditions, which are information relating to travel speed, road surface state, window open / close state, air conditioner operation state, output volume of received voice set in the speaker 12, and the like. These pieces of information are acquired from, for example, a navigation device (not shown) that is an in-vehicle device connected to the in-vehicle hands-free call level estimation device 3, or acquired from an ECU (Electronic Control Unit) that controls the vehicle body side. Or What is necessary is just to acquire the output volume set to the speaker 12 from the vehicle-mounted hands-free apparatus 1. FIG. The acquired vehicle conditions are output to the hands-free performance estimation unit 32 and the deterioration level improvement unit 34. Various vehicle conditions are conceivable, not limited to the above-described traveling speed and the like. In short, it is only necessary to indicate the state of an element that deteriorates the voice heard by the receiver during a hands-free call.

  The hands-free performance estimation unit 32 includes a sound including the user's voice output by the microphone 11 as an electrical signal and processed by the hands-free voice processing unit 15, and a vehicle condition acquired and output by the vehicle condition acquisition unit 31. Are used to estimate the voice degradation level during a hands-free call. The estimation result is output to the notification unit 33 and the deterioration level improvement unit 34. The deterioration level of the voice is a deterioration level of the voice of the user that the listener hears. Further, the sound processed by the hands-free sound processing unit 15 used for estimating the deterioration level is directly acquired from the hands-free sound processing unit 15, for example.

The notification unit 33 notifies the user of the voice deterioration level by outputting information on the voice deterioration level estimated by the hands-free performance estimation unit 32 to an output unit such as a monitor (not shown). The user may be notified by voice through the speaker 12.
The deterioration level improvement unit 34 is suitable for use in the hands-free sound processing unit 15 using the sound deterioration level estimated by the hands-free performance estimation unit 32 and the vehicle conditions acquired by the vehicle condition acquisition unit 31. The parameter is calculated, and the parameter stored in the parameter storage unit 16 is changed as necessary. Thereby, the voice processing by the hands-free voice processing unit 15 is changed.

As shown in FIG. 2, each unit constituting the in-vehicle hands-free call level estimation device 3 is realized, for example, when the processor 4 executes a program stored in the memory 5. Further, a plurality of processors and a plurality of memories may be linked to realize.
FIG. 1 shows a case where the in-vehicle hands-free call level estimation device 3 is configured as a separate device from the in-vehicle hands-free device 1. However, it may be configured integrally with the in-vehicle hands-free device 1. Further, since the in-vehicle hands-free device 1 is often configured integrally with a navigation device that performs route guidance or the like, the in-vehicle hands-free call level estimating device 3 may also be configured integrally with the navigation device.

Next, an example of the processing of the in-vehicle hands-free call level estimation device 3 configured as described above will be described using the flowchart shown in FIG.
During a hands-free call, the microphone 11 collects the voice uttered by the user. At this time, noises other than the user's voice such as a car running sound, an air conditioner blowing sound, and a received voice are also collected. The hands-free performance estimation unit 32 acquires the sounds collected by the microphone 11 and subjected to the voice processing by the hands-free voice processing unit 15 from the hands-free voice processing unit 15 (step ST1).
Subsequently, the vehicle condition acquisition unit 31 acquires vehicle conditions (step ST2).

Subsequently, the hands-free performance estimation unit 32 uses the sound acquired in step ST1 and the vehicle conditions acquired by the vehicle condition acquisition unit 31 to estimate the voice degradation level during a hands-free call (step ST3). ).
For the estimation of the deterioration level, the S / N ratio, the evaluation value of the voice quality, etc. calculated using the sound acquired in step ST1 are used. The smaller the S / N ratio, the higher the degradation level is estimated. As the voice quality evaluation value, for example, a sound quality evaluation index for an in-vehicle hands-free device formulated by the German Automobile Manufacturers Association is used. An example of such a sound quality evaluation index is TMOS (Tosqa Mean Opinion Score).

Vehicle conditions are also used to estimate the deterioration level. For example, the higher the traveling speed, the higher the level of deterioration is estimated. Moreover, it is estimated that a deterioration level is so high that the unevenness | corrugation of a road surface state is severe. Moreover, it is estimated that a deterioration level is so high that a window opening is large. In addition, the higher the output volume of the received voice set for the speaker 12, the higher the degradation level is estimated.
Furthermore, the hands-free performance estimation unit 32 may estimate the deterioration level in a higher direction as the air volume and the rainfall are larger, using the air volume and the rainfall information. The air volume and rainfall information can be acquired from an external server (not shown), for example. Moreover, you may obtain the information which can estimate the degradation level of an audio | voice for every traveling point from this external server. For example, the S / N ratio for each traveling point, an evaluation value of voice quality, a radio wave level, and the like. For example, the S / N ratio, the evaluation value of the voice quality, the radio wave level, etc. at each travel point are transmitted to and stored in an external server each time the host vehicle or another vehicle travels. If a travel point of a vehicle is acquired from a navigation device (not shown), the S / N ratio, the voice quality evaluation value, the radio wave level, and the like of the travel point can be searched from information from an external server. Using the search result, the hands-free performance estimation unit 32 estimates the deterioration level.

  As described above, the deterioration level can be estimated from various items such as the S / N ratio, the traveling speed, and the open / closed state of the window. The deterioration level may be estimated using only one of these items, or the deterioration level may be estimated comprehensively by combining a plurality of items.

Subsequently, the notification unit 33 notifies the user of information related to the voice degradation level estimated by the hands-free performance estimation unit 32 by screen display, voice output, or the like (step ST4). For example, the deterioration level is notified by an index of high, medium, and low, or the deterioration level is notified by a numerical index such as a five-step evaluation.
Thereby, the user can know what kind of sound quality has reached the receiver. When it is notified that the degradation level of the voice heard by the listener is high, the user can take improvement actions such as stopping or closing the window as a trigger. On the other hand, when notified that the deterioration level is low, the user can continue the hands-free call under the same conditions without anxiety.

In the flowchart of FIG. 3, the process for improving the deterioration level is continuously shown after the deterioration level is notified. This will be described below along steps ST5 to ST7.
The deterioration level improvement unit 34 uses the sound deterioration level estimated by the hands-free performance estimation unit 32 and the vehicle conditions acquired by the vehicle condition acquisition unit 31 to be used in the sound processing of the hands-free sound processing unit 15. A parameter that can be changed is output to the notification unit 33 (step ST5).
For example, when the output volume of the received voice is large, an echo cancellation parameter value that can appropriately reduce the echo of the received voice is obtained. Further, when the traveling speed is fast or the window is wide open, a noise cancellation parameter value that can appropriately reduce noise due to traveling noise is obtained. In addition, the noise caused by airflow from the air conditioner appears as a high frequency sound compared to the running noise. Therefore, the type of noise having a high deterioration level is estimated from the vehicle conditions, and an appropriate noise cancellation parameter value is obtained for each frequency band.

  Next, for example, the notification unit 33 outputs and displays whether or not to change the voice processing parameter on a monitor integrated with a touch panel (not shown) in a state where the user can select the parameter, and changes the parameter. Accept (step ST6). For example, a change execution button and a change unnecessary button are displayed side by side on the touch panel.

  When the user selects to change the parameter (step ST6; YES), the deterioration level improving unit 34 receives an operation signal from a monitor (not shown) indicating that and stores the parameter stored in the parameter storage unit 16. The parameter is changed to the parameter calculated in step ST5 (step ST7).

  On the other hand, when the user selects that the parameter change is unnecessary (step ST6; NO), the voice processing is continuously performed with the parameter stored in the parameter storage unit 16.

  In step ST5, the deterioration level improving unit 34 specifies an element that increases the deterioration level from the vehicle conditions acquired from the vehicle condition acquiring unit 31 instead of parameter calculation, and in the subsequent step ST6, the notifying unit 33 However, you may perform the process which notifies a user of the improvement of the specified main element. For example, when it is found from the vehicle conditions that the window is wide open and the deterioration level can be improved by closing the window, the deterioration level improving unit 34 outputs the fact to the notification unit 33. Then, the notification unit 33 transmits a message indicating that the deterioration level is improved if the window is closed to the user through screen display, audio output, or the like. As a result, the user can take an improvement action such as closing the window. Even under vehicle conditions, the output volume of the received voice set on the speaker 12 may be set not only manually by the user, but also automatically by the deterioration level improving unit 34 as with the parameters. Further, in step ST5, the deterioration level improving unit 34 may perform a process of identifying an element whose deterioration level is increased together with the parameter calculation.

  In step ST5, the deterioration level improving unit 34 may calculate a plurality of appropriate parameters. In step ST6, the notification unit 33 notifies the user of a plurality of parameter candidates. The more the echo cancellation parameter value is changed in such a direction that the echo of the received voice can be reduced, the more unnatural the user's voice heard by the receiver. Similarly, the more the noise cancellation parameter value is changed in a direction in which noise due to running noise can be reduced, the more unnatural the user's voice the listener hears. Because of such a phenomenon, a plurality of parameter candidates are notified so that the user can select the balance between the strength of the voice processing performed by the hands-free voice processing unit 15 and the naturalness of the voice. For example, if three candidates are to be notified, a parameter that produces unnatural speech with little noise and echo, a parameter that produces natural speech with much noise and echo, and an intermediate parameter between the two are assigned. The button is displayed in a selectable manner on a touch panel (not shown).

  Further, when the degradation level improvement unit 34 identifies an element that clearly increases the degradation level based on the vehicle conditions, the degradation level improvement unit 34 outputs the fact to the notification unit 33, and the notification unit 33 notifies the degradation level information. Instead, notification that the deterioration level is improved may be made by changing the state of the element. For example, when the traveling speed is slow and the air conditioner is stopped, but the window is wide open, the deterioration level improvement unit 34 immediately notifies the notification unit 33 that the deterioration level is improved by closing the window. To notify.

  In addition, when the vehicle condition acquisition unit 31 determines that none of the vehicle conditions is in a state of increasing the deterioration level, the vehicle condition is not used for subsequent processing. That is, the vehicle condition acquisition unit 31 estimates the hands-free performance. The vehicle conditions may not be output to the unit 32 or the like.

Further, the vehicle condition acquisition unit 31 may acquire vehicle conditions from the voice communication device 2. For example, the voice communication device 2 and a navigation device (not shown) are connected so that they can communicate with each other via Bluetooth, and the vehicle conditions handled by the navigation device are captured by the voice communication device 2 to obtain the vehicle conditions from the voice communication device 2. The unit 31 acquires vehicle conditions. Moreover, a navigation apparatus may send vehicle conditions to an external server, and the vehicle condition acquisition part 31 may acquire a vehicle condition from the said external server. That is, there are various vehicle condition acquisition routes.
Similarly, the parameters used by the hands-free voice processing unit 15 are not stored in the parameter storage unit 16 of the in-vehicle hands-free device 1 but are stored in the voice communication device 2 to be used in the in-vehicle hands-free call level estimation device. 3 and the voice communication device 2 may be connected in a state where they can communicate with each other via Bluetooth or the like, and may be configured to be changed by the deterioration level improvement unit 34.

  Further, the deterioration level improving unit 34 automatically selects the parameter used by the hands-free audio processing unit 15 and whether or not to change the output volume of the received voice set in the speaker 12 or not. The process of changing these settings may be performed.

  As described above, according to the in-vehicle hands-free call level estimation device according to the first embodiment, the sound collected by the microphone 11 and subjected to the voice processing by the hands-free voice processing unit 15, and the vehicle condition acquisition unit Using the vehicle conditions acquired by 31, the hands-free performance estimation unit 32 estimates the voice degradation level during a hands-free call. Then, the notification unit 33 notifies the user of the deterioration level estimated by the hands-free performance estimation unit 32. Thereby, the user can know the voice deterioration level in the hands-free call.

  Moreover, the deterioration level improvement part 34 which calculates a parameter suitable to be used by voice processing using the deterioration level estimated by the hands-free performance estimation part 32 and the vehicle condition acquired by the vehicle condition acquisition part 31 is provided. It was decided. Therefore, it is possible to perform appropriate sound processing according to the situation.

  In addition, the notification unit 33 outputs selectably whether or not the parameter used in the audio processing needs to be changed. When the user selects that the change is necessary, the deterioration level improvement unit 34 calculates the parameter used in the audio processing. It was decided to change to parameters. Therefore, the user can easily make the voice heard by the receiver have a low degradation level as needed.

  In addition, the vehicle condition acquisition unit 31 uses the vehicle condition acquired, and the deterioration level improvement unit 34 that outputs information indicating an action that can improve the deterioration level to the notification unit 33 is provided. Therefore, even if there is a limit to the improvement of the deterioration level only by the voice processing, the improvement of the deterioration level is expected.

  In addition, the deterioration level improving unit 34 calculates a plurality of parameters suitable for use in audio processing. Therefore, the user can appropriately select the state of the voice heard by the receiver.

  In the present invention, any constituent element of the embodiment can be modified or any constituent element of the embodiment can be omitted within the scope of the invention.

  DESCRIPTION OF SYMBOLS 1 In-vehicle hands-free device, 2 Voice communication device, 3 In-vehicle hands-free call level estimation device, 4 Processor, 5 Memory, 11 Microphone, 12 Speaker, 13 Control part, 14 Voice communication interface part, 15 Hands-free voice processing part, 16 Parameter storage unit, 31 vehicle condition acquisition unit, 32 hands-free performance estimation unit, 33 notification unit, 34 deterioration level improvement unit.

Claims (7)

A vehicle condition acquisition unit that acquires a vehicle condition indicating a state of an element that deteriorates a voice heard by a listener during a hands-free call;
Hands that estimate the deterioration level of the voice heard by the listener during a hands-free call using the sound that includes at least the voice uttered by the user and subjected to voice processing and the vehicle conditions acquired by the vehicle condition acquisition unit A free performance estimator,
An in-vehicle hands-free call level estimation device comprising: a notification unit that outputs the deterioration level estimated by the hands-free performance estimation unit.   A deterioration level improvement unit that calculates a parameter suitable for use in the voice processing using the deterioration level estimated by the hands-free performance estimation unit and the vehicle condition acquired by the vehicle condition acquisition unit; The in-vehicle hands-free call level estimation device according to claim 1. The notification unit outputs a selectable whether or not to change a parameter used in the voice processing,
The in-vehicle hands-free call level estimation device according to claim 2, wherein, when the user selects that the change is necessary, the deterioration level improvement unit changes a parameter used in the voice processing to a calculated parameter.   The vehicle-mounted hands according to claim 1, further comprising: a deterioration level improvement unit that outputs information indicating an action capable of improving the deterioration level to the notification unit using the vehicle condition acquired by the vehicle condition acquisition unit. Free call level estimation device.   The in-vehicle hands-free call level estimating device according to claim 2, wherein the deterioration level improving unit calculates a plurality of parameters suitable for use in the voice processing.   A navigation device comprising the in-vehicle hands-free call level estimation device according to any one of claims 1 to 5.   6. An in-vehicle hands-free device comprising the in-vehicle hands-free call level estimation device according to any one of claims 1 to 5.

Images