WO2019029869A1

WO2019029869A1 - Method for assigning an identity to a portable device

Info

Publication number: WO2019029869A1
Application number: PCT/EP2018/065583
Authority: WO
Inventors: Christoph Voigt
Original assignee: Audi Ag
Priority date: 2017-08-08
Filing date: 2018-06-13
Publication date: 2019-02-14
Also published as: DE102017213846A1

Abstract

The present invention relates to a method for assigning an identity of at least one speaker to at least one portable device situated in a vehicle, according to which method voice recognition is carried out on the vehicle-side to determine a voice signature and this signature is compared with a voice signature stored in an external server unit.

Description

A method of associating an identity with a portable device

The present invention relates to a method for assigning an identity of at least one speaker to at least one in one

Vehicle arranged portable device in which a vehicle-side voice recognition is performed for determining a voice signature and compared with stored in an external server unit voice signature.

In a vehicle, different functions can be implemented if the identity of the driver or a passenger can be automatically recognized. For example, by identifying the speaker in an interior of the vehicle, a portable device coupled to the vehicle may be assigned to the speaker and predetermined functions performed.

DE 10 2013 007 256 A1 discloses a method for the personalized operation of a graphic user interface of a motor vehicle and a motor vehicle with such a graphical user interface. The identity of a user of the motor vehicle is determined by a voice recognition.

DE 10 2013 013 539 A1 discloses a driver assistance system and a method for operating a driver assistance system. The

Driver assistance system has a biometric

Person recognition device with a voice recognition unit.

DE 10 2006 052 481 A1 discloses a method and a device for operating a vehicle with at least one driver assistance system. An identification of a driver is carried out using a driver voice sample. However, in the already known methods, no additional assignment of the identity of the speaker in the interior of a vehicle to a portable device takes place. Through this assignment can more

personalized functions, such as playing a speaker's preferred music or automatically making individual vehicle functions available. In addition, to carry out speech recognition and matching of a voice signature in the already known methods, a high computing power is necessary.

The invention is therefore based on the object, a method for

To provide a speaker in the fast and efficient way

Identify the interior of a vehicle and assign a arranged in the interior of a vehicle portable device to the speaker.

An object of the invention is a method for assigning an identity of at least one speaker to at least one in one

Vehicle arranged portable device. In this case, speech recognition is performed in an interior of a vehicle. A speech sequence is recorded and defined speech features are determined from the speech sequence. From the determined language features can a

Language signature can be created. Subsequently, the speech signature of the at least one speaker is compared with at least one stored in advance in an external server unit speech signature. at

If the created voice signature conforms to the at least one voice signature stored in advance in the external server unit, the at least one speaker is assigned one of the at least one voice signature assigned to the external memory unit and also stored in the external memory unit, wherein the assigned identity and thus the speaker is in turn assigned to the portable device.

The external storage unit comprises in an embodiment a database, which in turn comprises a number of language signatures and these unique identities. As a rule, the database comprises a multiplicity of language signatures and correspondingly a multiplicity of identities uniquely assigned to the respective language signatures. The wearable device may be a smartphone, a tablet, a smartwatch, and the like. The driver can use software, especially one

Application software, d. H. install a so-called app on his portable device and log on to the external server unit. By registering and storing personal data on the external server unit, the driver's identity can be stored in the portable device and on the external server unit. Preferably, the driver's identity in the external server unit may also be assigned to a registered driver's vehicle. Here, for example, further information, such as name and address, be deposited for an emergency.

The application software or app can be a variety of different software. The app is usually installed in the operating system of the portable device, which is then a native app. If the app is accessed via a foreign server, it is usually a web application. In this case, the app can be accessed via an interface installed, for example, on the portable device. Such a web app is stored on a remote server or web server, such as. In a cloud and is provided via the Internet via a browser interface, for example. The portable device. Mandatory for using a web app or application is an internet connection. Alternatively, it is also conceivable that the app is designed as a hybrid app. Hybrid apps are often used in conjunction with mobile devices. Here, a first part of the app, similar to a native app, on a local computer unit, in the present case on the portable device, and a second part of the app, similar to a web app, implemented on a remote server.

App stands as an abbreviation for the English word "application" (application). An application is a software program intended to perform a specific function directly to the user or, in some cases, to another application program. An app is typically a small piece of computer or mobile software that performs a specific function for a user. The app can be accessed locally by installing the software on the local portable device, such as a smartphone or tablet, which is then a desktop application.

Through the software or the app, the smartphone owner or driver or a passenger can deposit a unique voice signature in the portable device. The voice signature can be transmitted to the external server unit or a back-end system and automatically assigned to at least one vehicle of the smartphone owner or driver.

When using the speech recognition in the vehicle, the detected in the vehicle language signature with the deposited for this vehicle

Language signature in the external server unit to be matched. By the method according to the invention, for example, the driver can register easily and comfortably in the vehicle. The identity of the driver can be determined by a speech recognition performed in the portable device, followed by matching with the external server unit. In addition, the method can be used to reliably determine the portable device of the driver or a passenger by means of the voice, whereby, for example, a manual smartphone selection process for a driver

Connection with the vehicle may be redundant. As a result, the comfort for the occupants of the vehicle can be increased.

In one embodiment, speech recognition is based on

biometric speech features. For example, a defined voice sequence can be recorded. From the recorded speech sequence, individual speech features can be extracted. Alternatively, the speech features may be extracted from a text freely spoken by the speaker. Based on the speech characteristics, a speech sequence can be assigned to an individual speaker. The individual speech features may be, for example, in the external

Server unit and / or stored in the portable device. The speech features may be, for example, an accentuation sequence, word-pause patterns, and the like. By re-entering speech, the extracted speech features of the re-speech input with the

stored speech characteristics are compared. This can be an authentication of a person, such as a driver or

Passenger, by means of which biometric language features are carried out. A manual authentication process, such as by entering a user and a password, can thus be omitted. According to another embodiment, the speech recognition is in the

Interior of the vehicle by at least one installed on the at least one portable device and implemented software performed. For this purpose, the software may request, for example, a defined voice sequence or a freely spoken text. An input of the driver or any person in the interior of the vehicle can then by pronouncing a defined speech sequence or any free text. From this, speech features defined by the portable device can be extracted for subsequent comparison.

According to a further embodiment, the speech recognition in the interior of the vehicle is performed by at least one device of the vehicle. Alternatively or additionally, the

Speech recognition or authentication of a person by a microphone arrangement and corresponding software connection of a

Control unit of the vehicle done. For this purpose, the vehicle can

Record voice sequence and the relevant information for

Identify a person, such as name and

Vehicle identification number, determine. This can be done in conjunction with the portable device. For example, a microphone assembly of the vehicle may be used by the portable device via a USB connection or a Bluetooth connection to record the voice sequence. The subsequent speech sequence may be by the portable device or the external server unit.

According to a further embodiment, the voice signature of at least one speaker is stored in the external server unit.

The language signature with the extracted speech features can

preferably be stored in the external server unit.

The voice signature can be used independently of the vehicle. In addition, the speech recognition, regardless of the portable device, for example, via built-in interior of the vehicle devices, such as microphones or hands-free, done. Furthermore, by using different portable devices, an identification of the speaker can be made. It can be more complex Algorithms used for matching, as in the external

Server unit higher computing power can be realized.

According to a further embodiment, the voice signature of at least one speaker is stored in the external server unit via the software of the portable device. The software of the portable device can communicate with the external server unit. Through the communication, as part of a first authentication process, a copy of the language signature with the language features can be stored in the external server unit. The deposit can also be part of a

Backup copy done.

According to a further embodiment, the speech recognition in the interior of the vehicle is carried out continuously or situation-dependent. Speech recognition may be requested, for example, at a vehicle start or before vehicle start. Alternatively or additionally, based on conversations in the interior of the vehicle, it can be checked continuously or at regular intervals whether the driver's speech characteristics match the stored speech signature. This can be used for background authentication

be carried out, which requires no separate input of the driver.

According to a further embodiment, the adjustment of the at least one voice signature of at least one speaker by the software of the at least one portable device with the stored in the external server unit voice signature is performed continuously or situation dependent. Regardless of the interior of the vehicle

Speech recognition can be compared with the stored in the external server unit language signature. This Alignment can prevent, for example, that a vehicle start can be performed with unauthorized modified portable devices.

According to a further embodiment, the speaker is assigned to a portable device by assigning the identity of the at least one speaker to a voice signature. At least one portable device can be stored in the external server unit. This can for example be based on the digital identification of the portable device, such as a hardware identification. Through the determined language signature, the recorded speech sequence and the extracted

Speech features associated with the portable device and stored accordingly retrievable. This can be done for example in the external server unit. In this way, based on the determined

Voice signature automatically activated and controlled by multiple portable devices. For example, based on the recognized

Voice signature only the driver's smartphone is activated and a function performed by this smartphone.

According to a further embodiment, a position of at least one speaker in the interior of the vehicle is determined by the speech recognition. In this case, the position of the speaker can be reliably determined by a corresponding localization algorithm each time the voice control is used via voice processing in the vehicle. This can be done, for example, by the recognition of a speaker nearest to the microphone. For this purpose, for example, in the case of an arrangement of a plurality of microphones based on the determined level of different microphones of a recorded speech sequence, a sitting position of the speaker can be determined. According to another embodiment, the identity of at least one speaker is in one for the app installed on the portable device

accessible storage unit of the portable device and / or stored in the external server unit. Preferably, a speech sequence can be assigned to a speaker. The linking of the significant

Speech sequence with the extracted speech features with the identity of the corresponding speaker can allow a unique identification. In this way, an authentication of the speaker can be done quickly and without cumbersome manual input of access data.

According to a further embodiment, an identity of at least one portable device is stored together with the identity of at least one speaker in the external server unit. In addition to the biometric signature in the external server unit, the portable device associated with the driver can be stored. For example, the International Mobile Equipment Identity (IMEI) number may be used to associate a portable device. This makes it possible to limit the comparison of the biometric language signature to a few combinations. Only the vehicle, the portable device and the

Speech recognition are compared. This can speed up the alignment of the driver's data with the external server unit.

According to another embodiment, based on the identity of the at least one speaker and the determined position of the at least one speaker at least one custom setting of

Vehicle made. As a result, after the recognition of the identity of a person and the determination of the seating position in the vehicle deposited and person-specific preferences different

Settings are retrieved in the vehicle and made automatically. In this case, for example, person-specific preferences regarding a suspension tuning or a preferred one

Switching behavior are deposited. Other preferences could be

For example, affect settings of an air conditioning, a setting of a seat heating, a specific seat adjustment or a massage function. It can also be person specific music data or

Address book entries are stored in the vehicle. Furthermore, can

automatically accessing corresponding data of the portable device of an identified speaker. Once the identity and seating position of a speaker has been identified, the stored preferred settings can be recalled and made at the appropriate seating position. This eliminates a often consuming, manual adjustment and the comfort of the passengers and the driver can be increased. According to another embodiment, the software for the at least one portable device is provided by the external server unit.

Preferably, the software may be from the external server unit

be downloaded. For example, the external server unit may provide an app for the portable device. The software may interfere with the connection of the portable device to the vehicle through the external

Server unit are offered. This allows the software to be conveniently provided.

The invention is illustrated schematically with reference to an embodiment in the drawing and will be further described with reference to the drawing.

Figure 1 shows a schematic representation of an embodiment of the method according to the invention. Figure 1 shows a schematic representation of a method 1 for

Assigning an identity of at least one speaker according to an embodiment of the invention. Through the at least one speaker is a voice input 2. The voice input 2 is in accordance with the

Embodiment a defined pronounced sentence. The defined sentence thus represents a speech sequence. Alternatively, any sentence or phrase spoken during a conversation may be considered as

Speech sequence can be used. The voice input 2 is recorded by at least one microphone 6 arranged in a vehicle 4. Subsequently, the

recorded voice sequence to the portable device 8 forwarded.

For this purpose, the portable device 8 may preferably be coupled to the vehicle 4. According to the exemplary embodiment, the portable device 8 is a smartphone 8, which is connected or coupled to the vehicle 4 via a Bluetooth connection. The at least one microphone 6 may in this case be a hands-free device of the vehicle 4.

Based on the transmitted voice sequence is by the portable device 8 or by a software implemented thereon a

Speech recognition 10 performed. For this purpose, biometric speech features are extracted from the speech sequence by the portable device 8 and a biometric speech signature 12 is determined or created. The voice signature 12 detected by the portable device 8 becomes

subsequently compared with a number of biometric language signatures 16 stored in an external server unit 14. For this purpose, the portable device 8 can communicate with the external server unit 14 via a wireless Internet connection. Each voice signature 16 stored in the external server unit 14 is associated with an identity 18 of a respective speaker. A respective identity 18 of a respective speaker may include information about the person of the respective speaker, such as name and address, as well as information about a vehicle. In this case, in particular, a vehicle identification number and technical data of the vehicle can be stored. Furthermore, the smartphone used by a respective speaker of the respective biometric

Be associated with language signature. On the basis of the determined and compared speech signature 12, 16 the speaker can be assigned an identity. As a result, for example, even with a plurality of portable devices 8 arranged in an interior of the vehicle 4, the portable device 8 can be automatically assigned to the speaker on the basis of his voice signature 12, 16. In this way, for example, individual address books can be called up or messages can be sent via the smartphone of the speaker, even if there are several portable devices arranged in the vehicle.

In the external server unit 14 is usually a variety of

different identities 18 to different vehicles 4 and portable devices 8 deposited. In particular, a plurality of portable devices 8 and multiple identities 18 may also be associated with a vehicle 4. Also, multiple vehicles 4 may be assigned an identity 18, for example, if a driver has multiple vehicles 4.

LIST OF REFERENCE NUMBERS

1 procedure

2 voice input

4 vehicle

6 microphone

8 portable device

10 speech recognition

12 language signature

14 external server unit

16 stored language signature

18 deposited identity of a speaker

Claims

claims

1 . Method (1) for assigning an identity of at least one

A speaker for at least one portable device (8) arranged in a vehicle (4), comprising the steps:

- Performing a speech recognition (10) in an interior of a vehicle (4) and determining at least one voice signature (12) of at least one speaker;

Matching the at least one language signature (12) of the

at least one speaker with one in an external one

Server unit (14) stored language signature (16);

Assign an identity (18) of a speaker based on the

Language signature (12, 16) to at least one portable device (8) by at least one on the at least one portable

Device (8) implemented software.

2. The method of claim 1, wherein the speech recognition (10) is performed on the basis of biometric speech features.

3. The method of claim 1 or 2, wherein the speech recognition (10) in the interior of the vehicle (4) by the at least one software of the at least one portable device (8) and / or by at least one device of the vehicle (4) is performed ,

4. The method according to any one of claims 1 to 3, wherein the voice signature (16) of at least one speaker in the external server unit (14) is deposited.

5. The method according to any one of claims 1 to 4, wherein the adjustment of the at least one voice signature (12) of at least one speaker by the software of the at least one portable device (4) with the stored in the external server unit (14) speech signature (16) continuously or depending on the situation.

6. The method according to any one of claims 1 to 5, wherein by the

Voice recognition (10) a position of the at least one speaker in the interior of the vehicle (4) is determined.

Method according to one of claims 1 to 6, wherein the identity (18) of the at least one speaker is stored in the software of the portable device (8) and / or in the external server unit (14).

8. The method according to any one of claims 1 to 7, wherein an identity of at least one portable device (8) is stored together with the identity (18) of at least one speaker in the external server unit (14).

9. Method according to one of claims 1 to 8, wherein based on the identity (18) of the at least one speaker and the determined position of the at least one speaker at least one custom

Adjustment of the vehicle (4) is made.

The method of any one of claims 1 to 9, wherein the software for the at least one portable device (8) is provided by the external server unit (14).