DESCRIPTION:

The invention relates to a method for operating a telephone communication robot comprising an interactive voice response system. The invention also relates to a telephone communication robot and a vehicle comprising such a telephone communication robot.

It is known from the prior art that for example call center employees are replaced by robots, i.e. software with speech synthesis and dialogue system, which can then take calls and handle a customer request completely independently. Furthermore, it is known that such robots can also initiate calls, for example, to reserve a table in a restaurant for a customer or to carry out such a request like the smart assistant “Google Duplex”. Such robots use interactive voice response, which is a technology which allows humans to interact with such a computer operated phone system through the use of voice. Depending on how sophisticated such robots are, such a dialogue system may be restricted to a fixed dialogue structure or not. Especially intelligent assistants allow for restriction free conversation with counterparts, without the need of any predefined dialogue structure. Since such systems, especially telephone communication robots, save costs, and at the same time the state of technology behind them is becoming better and better, it can be assumed that such systems will become more and more popular. A great advantage of such telephone communication robots, which use natural language for communication, is that due to the use of such a natural language-based communication no specific interfaces have to be defined. This advantage, however, is 2 associated with a loss of efficiency. This is due to the fact that using a natural language for communication is much less efficient, e.g. compared with a communication between only computers using, for example, only structured requests defined by protocols.

US 2020/0065057 A1 describes a device, which allows a natural, humanlike conversation to be carried out between electronic apparatus. Thereby the device is configured as an audio adjustment device which is configured to adjust the conversation volume. This way the conversation between two robots may sound more natural in case a human is listening to the conversation.

DE 11 2018 000 702 B4 describes a robot system comprising a working robot and a dialogue robot, which is configured to interact with the working robot and to provide a dialogue to a human in form of speech.

It is the object of the present invention to provide a method, a telephone communication robot and a vehicle, which allow to provide a conversation over a telephone connection which is as efficient as possible.

This object is solved by a method, a telephone communication robot, and a vehicle, with the features according to the independent claims. Advantageous embodiments of the invention are presented in the dependent claims, the description, and the drawings.

According to a method for operating a first telephone communication robot according to the invention, the first communication robot comprises an interactive voice response system, wherein in case a telephone connection with a counterpart is established, the first communication robot determines whether the counterpart is a human being or a second communication robot, and in case the first telephone communication robot determines that the counterpart is a human being, the first telephone communication robot communicates with the human being by using a certain first language according to a set first communication speed, and in case the first telephone communication robot determines that the counterpart is a second communication robot, the first telephone communication robot communicates with the second telephone communication robot by using a certain second language according to a set second communication speed different from the first communication speed. 3

The invention is based on the finding that, as telephone robots are getting more and more popular, it can be assumed that one day, in a not insignificant number of cases, both, the caller and the recipient, will be a robot, for example, when a table is to be reserved in a restaurant out of a vehicle, wherein the restaurant also uses a robot to receive the calls. In this case, the driver advantageously does not have to “phone through” a number of restaurants himself but leaves the task to the digital assistant comprising the telephone communication robot. However, when using a language for the communication between two robots in an unmodified manner like in case of a communication between a robot and a human being, the communication is not as efficient as it could be. This is again based on the finding that in principle two robots can perform speech recognition and speech synthesis much faster than in real time, meaning much faster than the slow communication speed required for the communication with human beings. But now, according to the invention, the communication speed can advantageously be adapted depending on whether the first telephone communication robot communicates with a human being or another second telephone communication robot over the telephone. This allows for a much faster communication, when the counterpart is another telephone communication robot, even though a natural language is used for such a communication. As described in the beginning, using a language like a natural language for the communication has the great advantage that no interfaces have to be defined, especially for a fully automated communication, as is necessary today according to the state of the art. Usually, those advantages are associated with a loss of efficiency. However, this loss can be compensated advantageously by means of the invention. Moreover, efficient communication between two robots may be made downward compatible to the participation of a human participant, which saves resources in the development, as, in the example of restaurant reservation, no extra interface for automated requests has to be developed and provided. So, advantageously, with only minor technical changes, an efficient communication can be enabled for the case where both counterparts of a telephone call are robots. This is advantageous because, for example, in the case of a reservation, a dialogue, that may last several minutes, does not have to be carried out in such a long time, but can be completed much more quickly, so that, for example, the “rattling off” of various restaurants can be completed in just a few seconds. 4

A telephone connection can be established in form of a mobile communication network or a landline. The communication over the telephone connection does not make use of any additional data channels. In other words, when the first telephone communication robot communicates with the counterpart, the communication or conversation is fully performed by using the established telephone connection as single communication channel, and no additional channels for transmitting data between robots. Further, a telephone communication robot, like the first or second telephone communication robot, which comprise an interactive voice response system, can be provided in form of a software module. Such a software module can make use of known techniques of interactive voice response systems, including speech recognition and speech synthesis. The software module may also be implemented with an artificial neural network to provided more sophisticated communication dialogues, which especially need not be restricted to any dialogue structure. Nevertheless, also a fundamental dialogue structure can be used and be implemented. Moreover, the telephone connection can be established by dialing a certain telephone number. The first telephone communication robot can be the caller or the recipient receiving the call. In case, the first telephone communication robot, which in the following is simply called the first robot, is the caller, the first robot can also be configured to dial the defined telephone number, e.g. upon request by a user. Moreover, the second telephone communication robot, which is in the following simply called the second robot, can also be configured like the first robot.

According to another advantageous embodiment of the invention, the second communication speed is faster than the first communication speed, especially many times faster. So, advantageously, the communication over the telephone can be performed much faster and much more efficient between the two robots, especially in case no human being is involved. The first robot may comprise only two settings for the communication speed, like the first and the second communication speed, or even more different settings. Also, a continuous adaption of the communication speed could be possible. Providing more than two settings for the communication speed has the advantage that this allows for better adaptions to the situation and especially to the capabilities of the second robot. It may happen, for example, that the second communication speed is too fast for the second robot to perform adequate speech recognition. If this is the case, the first robot can adapt its communication speed again to be a bit slower and, for example, to communicate with a second robot according to a set 5 third communication speed being faster than the first communication speed but slower than the second communication speed. Also, in case the first robot is not capable of performing adequate speech recognition when communicating with the second robot, also the second robot may adjust the communication speed accordingly, e.g. upon request of the first robot.

Generally, the first language used for the communication with a human being, can be different from the second language used for the communication with a second robot. However, it is preferred that the first language is identical to the second language or at least does not depend on whether the first robot communicates with a human being or the second robot as the counterpart. For example, the language used for the communication can be set by a user of the first telephone communication robot, e.g., the user of a digital speech assistant comprising the first robot, for example, in a vehicle. The language can be selected from multiple different selectable languages. Further, the language can also be selected in dependency of the country, in which the first robot currently is or in which the recipient of the call currently is positioned. In other words, depending on whether the counterpart is a robot or a human being, preferably only the speed or speed parameters of the communication are adapted, but not the language of the communication.

According to another advantageous embodiment of the invention, the first language and the second language are a natural language or a constructed language, like Esperanto, but especially are not markup languages, like XML. This has the advantage that the same language can be used at the same time also for the communication with human beings, because such natural and constructed languages are understandable by human beings that are familiar with the specific language, and those languages are suitable for performing a conversation.

According to a very advantageous embodiment of the invention, for determining whether the counterpart is the human being or the second telephone communication robot, the first telephone communication robot determines, whether a specific first signal, which identifies the counterpart as being a robot, is received over the telephone connection, and in case the specific first signal is received, the counterpart is identified as the second communication robot. 6

So, advantageously the second robot may send a signal, here called the first signal, on the basis of which the second robot can be identified as a robot. Thus, the first robot recognizes that the counterpart is also a robot, and then the first robot can automatically switch to the second communication speed. Preferably, such a signal is sent at the beginning of a conversation, especially before the actual conversation, which is the purpose of the call, is started. Consequently, an efficient communication can be provided by adjusting the communication speed from the beginning.

Furthermore, according to another advantageous embodiment of the invention, when the telephone connection is established the first telephone communication robot sends a specific second signal for identifying the first telephone communication robot as a robot of the telephone connection, especially prior to starting the communication with a counterpart. Thus, also the first robot can identify itself as being a robot by sending the specific second signal. Thereby, the first and second signals can be the same or similar signals. Whether such a signal is sent by the first or second robot may also depend on whether the respective robot is the caller or the recipient receiving the call. For example, only the caller may send such a signal, and the recipient robot adjusts its communication speed accordingly. The caller who sent the signal then recognizes that the recipient is also a robot on the basis of the adjusted communication speed, according to which the recipient robot communicates over the telephone connection with the caller robot, or in the alternative on the basis of a response signal sent by the recipient robot. On the other hand, also only the recipient robot may send such a signal, on the basis of which then the caller robot adjusts its communication speed. Further, it is also possible that both, the caller and the recipient send such a signal to identify themselves as being robots. Sending such a signal is a very simple and efficient way of informing the counterpart about being a robot. At the same time, such a signal cannot not only be used to inform a robot as counterpart, but also a human being. This can also have advantages for the human being, because a human being would like to know whether he or she is talking to a robot or a real human being as counterpart.

Thus, when the first robot sends such a specific signal, the signal can be configured such that it is recognizable or receivable by a human being. Therefore, it is another advantageous embodiment of the invention when the first and/or second signal is 7 configured such that it is recognizable by a human being. As already explained, this has the advantage that such a signal can also be used to inform a human being as counterpart that he or she is taking to a robot.

In general, the specific first and/or second signal can be configured as a tone comprising only a certain frequency or a sequence of tones, each comprising a certain frequency, or a sound comprising one or more certain frequencies or a sequence of sounds each comprising one or more certain frequencies, or a word or expression or sentence, especially in the selected language. So, the signal can be, for example, a standardized tone or a natural language announcement, by means of which a human being as well as the second robot can identify the first robot as being a robot. So, also a human caller or a recipient would therefore know that he is currently taking to a robot. An example of such an announcement or expression will be “I am a robot” or similar. Using such a natural language-based signal in form of a word, expression or sentence, also no adaption of generating or recognizing such a signal is necessary. Thereby, the word, expression or sentence can be a defined key word, key expression or key sentence, but need not. That the counterpart is a robot can also be derived from the content of such word, expression or sentence by a speech recognition algorithms without defining those as specific key elements in advance.

As already mentioned, such a signal preferably is sent prior to the communication or at the beginning of a communication. For example, immediately after a start of the call, the called robot sends a signal that identifies it as a robot. This can be, for example, a tone on a certain frequency. The other system, that is the other robot, recognizes the signal and knows that it is a robot. Thereupon, it adapts the communication, by changing one or more parameters. For example, it could increase its own speech speed by a multiple and at the same time set parameters in the speech recognition to a high speech speed of the other station, namely the counterpart.

The invention also relates to a telephone communication robot comprising an interactive voice response system. Thereby, in case the telephone connection with the counterpart is established, the telephone communication robot is configured to determine whether the counterpart is a human being or a second communication robot, and the telephone communication robot is configured such that in case the telephone communication robot 8 determines that the counterpart is a human being, the telephone communication robot communicates with the human being by using a certain first language according to a set first communication speed, and in case the telephone communication robot determines that the counterpart is a second telephone communication robot, the telephone communication robot communicates with the second telephone communication robot by using a certain second language according to a set second communication speed different from the first communication speed.

The advantages described with regard to the method and its embodiments according to the invention similarly apply to the telephone communication robot according to the invention.

Moreover, the invention also relates to a vehicle comprising a telephone communication robot according to the invention or one of its embodiments.

The invention also comprises embodiments that provide features which afford additional technical advantages.

The invention also comprises embodiments of the inventive telephone communication robot that comprise features that correspond to features as they have already been described in connection with the embodiments of the inventive method. For this reason, the corresponding features of the embodiments of the inventive telephone communication robot are not described here again.

The inventive vehicle is preferably designed as a motor vehicle, in particular as a passenger vehicle or a truck, or as a bus or a motorcycle.

The invention also comprises the combinations of the features of the different embodiments.

In the following an exemplary implementation of the invention is described. The figures show in: 9

Fig. 1 a schematic illustration of a vehicle with an assistant system comprising a telephone robot according to an embodiment of the invention; and

Fig. 2 a schematic illustration of a flow chart for illustrating a method of operating a telephone communication robot according to an embodiment of the invention.

The embodiment explained in the following is a preferred embodiment of the invention. Flowever, in the embodiment, the described components of the embodiment each represent individual features of the invention which are to be considered independently of each other and which each develop the invention also independently of each other and thereby are also to be regarded as a component of the invention in individual manner or in another than the shown combination. Furthermore, the described embodiment can also be supplemented by further features of the invention already described.

In the figures identical reference signs indicate elements that provide the same function.

Fig. 1 shows a schematic illustration of a vehicle 10 comprising an assistant system 12 with a first communication robot 14 according to an embodiment of the invention. The assistant system 12 can assist a driver or user 16 of the vehicle 10 with a variety of tasks, one of which is automatically performing phone calls, for example, to make a reservation in a restaurant. For this purpose, the telephone communication robot 14 comprises an interactive voice response system 18. Such an interactive voice response system 18 can be configured as a software module 18. The software of such a software module 18 is executed by means of a processing unit of the first robot 14. Further, e.g., upon request of the user 16, the first robot 14 can establish a telephone connection 20 to a counterpart 22, e.g., by dialing a certain number associated with the counterpart 22. In this case, the telephone connection 20 is established in form of a mobile communication network. Such telephone connection 20 is in fact not a direct connection between the first robot 14 and the counterpart 22 but an indirect connection involving further communication elements like radio towers and/or satellites. The dashed line illustrating the telephone connection 20 is only drawn as direct line for the sake of simplicity. 10

When the call of the first robot 14 is answered by the counterpart 22, the first robot 14 may determine, whether the counterpart 22 is a human being or, as in this case, is a second telephone communication robot 24. This can be done in several ways, for example, immediately after the start of the call, the called robot 24 can send a signal S1 that identifies it as a robot. This can be, for example, a tone on a certain frequency. The other system, in this case the first robot 14, recognizes the signal and therefrom knows that the counterpart 22 is a robot. Also, the first robot 14 may send such a signal S2 to the counterpart 22. A human being as well as a robot 24 as counterpart 22 can therefrom derive the information that the call comes from a robot. Now, advantageously the first robot 14, as well as the second robot 24, can adapt communication parameters for the communication between each other. For example, the first robot 14 can set a parameter P to a first parameter value P1 in case the first robot 14 would have determined that the counterpart 22 is a human being. As the first robot 14 has determined, that the counterpart 22 is not a human being but also a robot 24, the first robot now sets this parameter P to a second parameter value P2. By adjusting the parameter P to the second value P2, the communication speed, which in the following is called speech speed, used for the communication over the telephone connection 20 can be increased in comparison to the speech speed used for communications with human counterparts. So, this speech speed, which would have been used in case the counterpart 22 would have been human, can be represented by the first parameter P1 . So, in this case, advantageously, the first robot 14 can increase its own speech speed according to the second parameter value P2, for example, by a multiple. Also, other parameters for the communication can be adapted. For example, at the same time parameters of the speech recognition can be sent and adapted to the high speech speed of the counterpart 22, which is also a robot 24.

In another embodiment, the calling system, here the first robot 14, sends a signal S2 and initiates the recognition. In another embodiment, both systems may send such a signal S1 , S2. Provided the signal is detected by one side, it may send an acknowledgement, this also identifying itself as a robot. Consequently, one of the signals S1 , S2 may be configured as such an acknowledgement signal or response signal.

Also, further signals, which can be send by the respective robots 14, 24, can be defined, like “please reformulate request” or “please reduce speed”. Such signals can be sent in 11 case speech signals of the respective counterparts cannot be recognized. Based on such signals, the robots 14, 24 may, for example, adapt the speech speed again and set it to some intermediate slower value. This slower value may still be much faster than the normal speech speed according to the first parameter P1 , which is used for human communication.

Though the communication between the two robots 14, 24 can be performed with increased speed, the robots 14, 24 still use a natural or constructed language for such a communication over the telephone connection 20. Using such a natural or constructed language has the great advantage that no special interfaces have to be defined. The interfaces used for the communication of human beings as counterparts are the same as used for the communication with other robots as counterparts.

Fig. 2 shows another schematic illustration of a method for operating a telephone communication robot according to an embodiment of the invention. The method starts in step S10, in which the first robot 14, for example, as explained with regard to Fig. 1 , dials a certain telephone number to establish a telephone connection 20 with a counterpart 22. After that in step S12, this telephone connection 20 is established. After that in step S14, the first robot 14 sends a specific signal S2 by means of which the first robot 14 is identifiable as being a robot. Such a signal S2 can also be a speech signal in the form of “I am a robot”, which is a speech signal comprising the content that the first robot 14 is a robot and not a human being. After that in step S16, the robot 14 determines whether the counterpart 22 is a human being or another robot 24. In case no such signal S1 is received, which would identify the counterpart 22 as being a robot and/or instead an answer from a human being is received that the counterpart 22 is a human being, than the method proceeds in step S18, in which the first robot 14 sets its communication parameter P to its first value P1 , which defines among others a normal, moderate speech speed for the communication with the counterpart 22, which is slow enough to be understandable by a human being. After that the first robot 14 communicates with the counterpart 22 in step S20, using a language and a speech speed according to the set first parameter P1. If on the other hand in step S16 a signal S1 is received from the counterpart 22, which identifies the counterpart 22 as being a robot 24, then, in step S22, the first robot 14 sets its parameter P to at least one second value P2, which defines a higher communication speed. Then again in step S20, the first robot 14 performs the 12 communication over the telephone connection 20 with the counterpart 22, but in this case according to the set second parameter P2.

To conclude, the examples show how by means of the invention an accelerated natural language communication between non-human partners can be provided in an easy and efficient way. Thus, with only minor technical changes, an efficient communication can be enabled for the case, where both parties are robots. This advantageous because, for example, in the case of a reservation, a dialogue that may last several minutes “how many people, when, name, etcetera” does not have to be carried out slowly, but can be completed much more quickly so that, for example, the “rattling off” of various restaurants an also other defined tasks can be completed in just a few seconds. The customer in the car can be given feedback within seconds such as that the restaurant has been reserved or three restaurants in the vicinity would be possible instead of minutes. Another advantage is the possible cost saving in development because no specific interfaces have to be defined and developed.

13

REFERENCE SIGNS:

10 vehicle

12 assistance system 14 first telephone communication robot

16 user 18 interactive voice response system 20 telephone connection 22 counterpart 24 second telephone communication robot

P parameter P1 first parameter value P2 second parameter value S1 first signal S2 second signal

S10 step S12 step S14 step S16 step S20 step S22 step