Voice control system, corresponding motorcycle, helmet and method

Abstract

Embodiments of the present disclosure relate to voice control systems, corresponding motorcycles, helmets, and methods. Embodiments include: a receiver circuit device that receives a voice-generated signal; a command recognition circuit device that recognizes a voice-generated command signal for a motorcycle dashboard from a voice-generated signal received at the receiver circuit device; and a command implementation circuit device that implements a motorcycle dashboard action based on the voice-generated command signal recognized by the command recognition circuit device. The command recognition circuit device includes a local command recognition circuit configured to recognize a voice-generated command signal and provide the locally recognized voice-generated command signal to the command implementation circuit device; a network connection interface configured to transmit the voice-generated signal to a network-based voice-generated command signal recognition device, receive the voice-generated signal from the network-based recognition device, and provide the command implementation circuit device with the voice-generated command signal recognized by the network-based signal recognition device.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the benefit of Italian Application No. 102020000000319, filed on January 10, 2020, the contents of which are incorporated herein by reference. Technical Field

[0003] The present description relates to voice control systems and methods, and in particular embodiments, to motorcycle voice control systems and methods. Background Art

[0004] An approach that is still widely popular in the motorcycle (motorcycle) market is to consider the motorcycle dashboard to be manually controlled: that is, the motorcycle rider uses one of his or her hands to control the dashboard, which almost inevitably involves temporarily loosening or at least changing the grip on the handlebars.

[0005] Although attractive and desirable, implementing a voice recognition system for motorcycle dashboard controls presents various problems and aspects to be addressed, such as:

[0006] High noise levels that may enter the microphone during operation,

[0007] high CPU load (which indicates the use of a powerful SoC),

[0008] It is expected that the battery life of hands-free devices equipped with motorcyclists' helmets may be reduced.

[0009] It may be related to the lack of network connectivity in some areas, with reduced access to premium services (or no access at all), and no backup solutions in the event of a lack of connectivity. Summary of the invention

[0010] It is an object of one or more embodiments to help resolve this problem, thereby facilitating implementation of a reliable voice recognition system for motorcycle dashboard controls.

[0011] According to one or more embodiments, this object can be achieved by means of a system having the functionality of the claims presented below.

[0012] One or more embodiments may relate to a corresponding motorcycle.

[0013] One or more embodiments may relate to a corresponding motorcyclist's (motorcycle driver's) helmet.

[0014] One or more embodiments may be directed to a corresponding method.

[0015] The claims are an integral part of the technical teaching provided with respect to the embodiments described by way of example herein.

[0016] One or more embodiments provide a HW / SW based solution that may include a "smart" helmet combined with a motorcycle onboard control unit (e.g., dashboard), capable of providing voice commands regardless of network online status, with the ability to use both local and cloud-based speech recognition engines, which may result in a cost-effective system.

[0017] One or more embodiments may provide a hybrid solution that facilitates operating such a control unit both with and without network availability.

[0018] In one or more embodiments, system operations may be triggered due to voice commands detected via a simple language detection process, which may reduce CPU load and, therefore, battery consumption on the smart helmet side.

[0019] One or more embodiments may rely on voice commands to render dashboard controls more easily (more user-friendly) while reducing CPU usage, which can result in increased battery life in a smart helmet.

[0020] In one or more embodiments, reduced CPU load may be associated with waking up and triggering the system since the first filtering action is intended to detect the presence of speech in noise.

[0021] For example, this may occur by running a speech detection process involving approximately 2 MIPS on a microcontroller or microprocessor device as described below.

[0022] Although primarily beneficial for helmet mounted arrangements, this may also be beneficial for control units mounted on the motorcycle (eg dashboard).

[0023] One or more embodiments may be effectively applied to both low-cost digital instrument control systems and high-end digital instrument control systems (eg, instrument clusters) for the motorcycle market.

[0024] One or more embodiments may facilitate access to cloud-based services in the presence of network availability, while also facilitating multi-user interaction.

[0025] One or more embodiments may provide a control system in which an onboard control unit (e.g., dashboard) of a motorcycle (as described below, the design is intended to also be applied to two-wheeled or three-wheeled vehicles such as scooters or mopeds) can receive voice commands from a user. As illustrated herein, the system may include a processor (e.g., a microcontroller) that communicates with headphones (also equipped with a wireless communication module) mounted on a user's helmet via one or more wireless communication modules (e.g., Bluetooth). The processor also communicates with a network (such as the Internet) via a wireless communication module (e.g., a smartphone) to perform functions involving cloud services.

[0026] Optionally, the system may also operate in conjunction with a motorcycle mounted microphone (eg, this may only operate when the motorcycle is not being driven / running).

[0027] In one or more embodiments, voice detection via a processor (e.g., a microcontroller / microprocessor) mounted on the motorcyclist's helmet may activate wireless transmission (only) when voice activity is detected.

[0028] In one or more embodiments, such a helmet mounted processor may perform a recognition process of one or more "wake-up words" and transmit to the motorcycle mounted processor (only) as a result of a positive result of such recognition process.

[0029] Therefore, based on such voice activity and / or recognition of a wake-up word processed at the helmet, one or more embodiments can avoid unnecessary transmission activity, thereby benefiting energy conservation and extending the life of the helmet's battery resources.

[0030] In one or more embodiments, voice command recognition (interpretation) may occur via a "local" process implemented in a motorcycle-mounted processor and / or via network (e.g., cloud) resources, with command interpretation potentially occurring locally in the absence of network availability.

[0031] In one or more embodiments, where there are available network resources, identification may be performed in two ways (eg, locally or via the cloud), and the first (earlier) identification result may be used for control purposes. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] One or more embodiments will now be described, by way of example only, with reference to the accompanying drawings, in which:

[0033] Figure 1 is an overall functional diagram of a system according to an embodiment of the present specification;

[0034] Figure 2 is a block diagram of an example of a possible embodiment according to the present specification;

[0035] Figure 3 is another block diagram of an example of a possible embodiment according to the present specification;

[0036] Figure 4 is yet another block diagram of an example of a possible embodiment according to the present specification;

[0037] Figure 5 is a block diagram of an example of possible variations that may be applied to the embodiments according to the present specification; and

[0038] Figures 6 to 8 is a block diagram showing a motorcycle processor acting as a bridge connecting a smart device in parallel with a motorcycle helmet. DETAILED DESCRIPTION

[0039] In the following description, various specific details are given to provide a thorough understanding of various exemplary embodiments of this specification. The embodiment can be practiced without one or more specific details or with other methods, parts, materials, etc. In other examples, known structures, materials or operations are not shown in detail to avoid confusing various aspects of the embodiment. References to "an embodiment" or "an embodiment" throughout the specification mean that specific functions, structures or features described in conjunction with the embodiment are included in at least one embodiment. Therefore, the phrase "an embodiment" or "an embodiment" may appear in various places throughout the specification, and does not necessarily all refer to the same embodiment. In addition, specific functions, structures or features can be combined in any appropriate manner in one or more embodiments.

[0040] The headings / references provided herein are for convenience only, and therefore do not interpret the scope or extent of the embodiments.

[0041] For example, one or more embodiments may be applied to a motorcycle voice control system.

[0042] As used herein, the name "motorcycle" will apply to a self-propelled two- or three-wheeled motor vehicle adapted to be ridden by a rider (a motorcyclist, usually wearing a helmet), without regard to the various variations that may exist in motorcycle design to suit a variety of different uses.

[0043] As used herein, regardless of whether a vehicle complies with currently applicable laws / regulations in a country or jurisdiction, the designation motorcycle applies. A scooter or a moped may be used as an example of a vehicle under the designation motorcycle as used herein.

[0044] Figure 1 is an overall exemplary view of a (hybrid) motorcycle voice control system 10 according to the embodiment illustrated herein.

[0045] like Figure 1 As illustrated in FIG. 1 , the system 10 comprises a processing circuit device 100 adapted and configured (in a manner known to those skilled in the art) to be mounted on a motorcycle (motorbike) MB, having the capability to communicate with:

[0046] at least one "smart" helmet SH worn by a motorcyclist riding the motorcycle MB: this may be done via a wireless module 200 (e.g. Bluetooth BT) with a corresponding module 200A of the equipped helmet SH; as discussed below, possible communication with another smart helmet SH2 worn by another motorcyclist of the motorcycle MB (possibly "tailored" to the voice control configured by another user of the motorcycle MB) or even a passenger may be considered in one or more embodiments;

[0047] The mobile communication device UE supports data communication and provides access to network resources such as Internet resources, for example (cloud data storage and computing as discussed below are examples of such resources); in one or more embodiments, the network connection can be made using a smart phone (the smart phone worn by the motorcyclist can be, for example, Figure 1 An example of such a device illustrated in UE 200 is implemented as a bridge cooperating with a wireless module (such as BT module 200 or a WiFi device).

[0048] In one or more embodiments, the operation of the system 100 as illustrated herein may be fully supported by processing circuitry 100, such as may be utilized using Accordion Processor, commercially available from STMicroelectronics. TM series of devices (see st.com).

[0049] It should also be understood that throughout this specification, references to Accordo TM References to devices are for exemplary purposes only: in fact, one or more embodiments lend themselves to implementation on a variety of microcontroller (MCU) / microprocessor (MPU) platforms currently available on the market.

[0050] Therefore, although it is helpful to understand the embodiments, it is not necessary to TM Repeated reference to devices should not be understood, even indirectly, as a limitation of the embodiments.

[0051] Accordo TM devices are a family of devices that provide cost-effective microprocessor solutions for existing car radio systems, including MIPS high-efficiency dual-core 32-bit ARM, including Cortex-R4 ( Figure 1 101) and Cortex-M3 ( Figure 1102) processor capability, the ARM Cortex-M3 controller 102 is dedicated to CAN ( Figure 1 Real-time vehicle interface processing 103) in the host device 100 can be connected to an electronic control unit (ECU) in a vehicle (such as a motorcycle or a motorcycle MB of the host device 100).

[0052] Embedded SRAM, audio ADC and DAC (see Figure 1 104 and 105 in, for example, having the ability to support the operation of sensors S (such as temperature, light or battery charge sensors), independent CAN MCU for car radio and display audio applications and embedded powerful digital sound processing ( Figure 1 The DSP-106) subsystem in the Accordo TM Further advantageous features of the series device.

[0053] like Figure 1 As exemplified in , these devices also include a detailed set of public interfaces, such as:

[0054] UART 107 and UART 108, capable of supporting line communication with handlebar buttons and joysticks at handlebars HB of motorcycle MB on line LIN,

[0055] I2S,109,

[0056] USB, 110

[0057] Video input interface 111,

[0058] The CLCD / TSC display interface 112 is, for example, configured to cooperate with a dashboard display unit D of the motorcycle MB via an associated 2-2.5D memory arrangement 112A,

[0059] The GPIO (General Purpose Input / Output) interface 113 can, for example, support the LED indicators TT (brake, low oil level) equipped on the motorcycle MB.

[0060] This helps to achieve a feature-rich system as well as a cost-effective solution supported by a complete software package allowing very fast system implementation.

[0061] Therefore, Accordo TM The devices in the series are examples of equipment suitable for managing the entire audio chain from analog or digital input to analog or digital output, including decoding of digital audio media, sample rate conversion between various sources, intelligent routing and audio effects / DSP post-processing, with the ability to support both low-cost systems based on real-time OS and to scale up to high-demand applications based on Linux OS.

[0062] In short, Accordo TM Devices (such as, for example, a combination of two devices, Accordo 2 and Accordo 5) can have display / graphics capabilities, media and vehicle connectivity capabilities, and audio functionality that facilitate their use in digital instrument applications for motorcycles (including the various types of 2- and 3-wheeled vehicles previously discussed).

[0063] This may be caused by features such as:

[0064] Ability to manage visual human-machine interface (HMI);

[0065] Connectivity to available mobile devices;

[0066] Vehicle interfaces suitable for connection to vehicles such as motorcycles;

[0067] (three) the availability of an integrated DSP, originally designed for audio effects and erasure coding (EC) and noise reduction (NR) acceleration;

[0068] The ability to receive sound input from a traditional helmet equipped with Bluetooth communication (already available on the market) and compatible with low-noise conditions, as well as the additional ability to receive sound input from a local microphone embedded in the motorcycle (such microphones are currently already used for safety purposes, such as facilitating emergency calls).

[0069] Figures 2 to 5 It is a combination Figure 1 Block diagram examples of various possible embodiments of the system 10 discussed, where the processing / operation functions are distributed to the units 100 in different ways (e.g., Accordo TM equipment).

[0070] In fact, in other embodiments, some of the processing / operational functions assigned to unit 100 may be assigned to other components, such as one or more ( Figure 4 )The smart helmets SH, SH2 are intended to be worn by motorcyclists riding the same motorcycle (such as MB) at different times in possibly different usage (control) configurations.

[0071] Figure 2 is an example of an embodiment, wherein the combination Figure 1 The processing / operating functions of the system 10 discussed are almost entirely distributed to the unit 100 (suitable for use with the previously discussed Accordo TM Device implementation), the unit 100 is installed in a manner known to those skilled in the art. Figure 1 On the motorcycle shown in ( Figures 2 to 5 (not visible in the image).

[0072] like Figure 2 The embodiment illustrated in is an example of a system 10, which is configured to operate in a "traditional" scenario in conjunction with a "smart" motorcyclist helmet SH equipped with a microphone MIC and / or a motorcycle-mounted microphone MIC2 (e.g., this can only be operated when the motorcycle is not driving / running), using the microphone MIC and / or microphone MIC2 in the smart helmet to collect sound messages, such as sounds emitted by the motorcyclist.

[0073] For example, as discussed above, such helmets equipped with microphone MIC and / or microphone MIC2 are currently already used for safety purposes. Therefore, they can be regarded as "traditional" equipment for motorcycles MB, which may be intended to be retrofitted with unit 100 as discussed herein in order to provide a voice control system according to an embodiment.

[0074] In such Figure 2 In the embodiment illustrated in FIG. 1 , the unit 100 may be equipped with a receiver module 200RX (eg, Bluetooth, see Figure 1 108, 109) to receive sound messages from the homologous transmitter module 200A in the microphone-equipped smart helmet SH and / or from the microphone MIC2. As mentioned above, such helmets SH with associated microphones MIC and transmitter / receiver modules (e.g., Bluetooth) are currently used, which makes it unnecessary to provide a more detailed description herein.

[0075] In one or more embodiments, regardless of whether the source is a microphone-equipped smart helmet SH or microphone MIC2, the voice message emitted by the motorcyclist can be provided to the voice activity detection module 302 at the (mixing) node as an audio stream.

[0076] Here, the audio stream may be analyzed by a voice activity detection process, thus facilitating a check (as illustrated by block 304 ) whether a voice signal is detected. In the latter case (negative result N from 304 ), the system re-loops onto the voice activity detection module 302 .

[0077] In one or more embodiments, the voice activity detection process may be of any type known to those skilled in the art that may provide power optimized operation that reduces CPU usage.

[0078] Block 306 (at 304 the system progresses due to a positive outcome Y) is an example of possibly using a wake word engine (again of a known type) to detect a keyword (e.g.: “Alexa,” “Siri,” “OK Google”) to initiate voice command processing.

[0079] The wake word processing at 306 facilitates checking (as illustrated by block 308) whether the wake word is detected. In the event that the wake word is not detected (negative result N from 306), the system progresses to the EXIT state.

[0080] Block 310 (system progresses due to positive result Y at 308) is an example of a check whether a network connection (eg, Internet) is currently available: this may be done via Bluetooth 200 and a device UE (eg, smartphone, hotspot, etc.).

[0081] As a result of the check for network availability performed at 310, the following request is sent:

[0082] In case of a negative result N, at 310 (network unavailable), a request is sent to a local speech recognition engine (eg, command interpreter) 312; and

[0083] In the case of a positive result Y, at 310 (network available), a request is sent to a network-based speech recognition engine (e.g., cloud command generator) 314.

[0084] In the former case (local engine), a check is performed at 316 whether the command is recognized.

[0085] A negative result N at 316 will lead the system to the EXIT state.

[0086] Conversely, a positive result Y at 316 will cause the corresponding action to be implemented (immediately). This can cause the control module 320 (e.g., Figure 1 112 and 113 in ), which results in the activation of certain desired functions at the display D or any LED indicator TT, possibly via the transmitter module 200TX (for example, such as a Bluetooth transmitter module).

[0087] Functions such as changing pages, showing range, showing battery charge level, activating audio to the helmet, to name a few, are examples of such possible activated actions.

[0088] In the latter case considered at 310 (network connection available), the audio command can be encoded at the cloud command generator 314 and can be used in Figure 1 The network connection illustrated in (eg, BT module 200, smart phone, hotspot) sends audio commands to a network-based command recognition resource (such as a cloud-based command recognition engine).

[0089] Thus, command recognition may be performed at the cloud level indicated by C, which may require more sophisticated recognition / classification capabilities than a local engine (such as 312).

[0090] Compared to a cloud-based engine, a local engine (such as 312) can address (only) a subset of commands, for example:

[0091] Cluster page management (move to diagnostic page, move to navigation page, move to rear view camera page, ...),

[0092] Activation of infotainment features (voice over phone call, received SMS or IM, ...).

[0093] By comparison, a cloud-based engine adapted to arrive via 314 and 200 may manage additional commands in addition to those managed by the local engine, such as:

[0094] Retrieve information (traffic, weather, news, ...),

[0095] Control by external entities,

[0096] Connection door management,

[0097] Smart home and other devices.

[0098] At 319, check that the network (e.g., cloud C, can again be connected via Figure 1 A response is received by the BT module 200 illustrated in FIG. 2 , a smart phone, a hotspot, etc. , as to whether the command is recognized at the cloud level C.

[0099] A negative result N at 319 will again direct the system to the EXIT state.

[0100] In contrast, a positive result Y at 319 will cause the corresponding action to be implemented (immediately).

[0101] This may again be via the converged processing node 318, control module 320, and transmitter module 200TX as discussed above, for example, with the proviso that more powerful command recognition at the network (cloud) level may result in a wider and more sophisticated range of actions being adapted for voice control.

[0102] Figure 3 and Figure 4 is an example of an embodiment, wherein Figure 2 Certain processing / operation functions assigned to unit 100 in the embodiment illustrated in the figure may be “moved” to other elements (such as one or more smart helmets SH, SH2).

[0103] Therefore, in Figure 3 and Figure 4 (as well as Figure 5 ) in combination with the components, elements or functions Figure 1 and Figure 2Similar components, elements or functions are described and will be indicated by similar reference numerals, and the description of these components, elements or functions will not be repeated for the sake of brevity.

[0104] Figure 3 is an example of an embodiment in which, due to the increased processing power at a smart helmet (such as SH) in a manner known to those skilled in the art, voice activity detection processing (e.g., involving about 2 MIPs) is performed on the helmet-level digitized voice signal at the helmet SH as again illustrated at 302 and 304 via an ADC converter 301, which provides digital conversion of the voice signal provided by the helmet-mounted microphone MIC (for simplicity, at Figure 3 and Figure 4 It is assumed that there is a single helmet-mounted microphone, in which the microphone MIC2 that may be mounted on the motorcycle is not visible).

[0105] This approach may prove advantageous in reducing helmet battery consumption, since an unnecessary continuous signal flow (e.g., BT) from the helmet transmitter 200A to the receiver 200RX in the unit 100 is avoided: in fact, only audio detected as speech (result of 304 = Y) may flow from the transmitter 200A to the receiver 200RX.

[0106] In such Figure 3 In the embodiment illustrated in FIG. 1 , the first process performed in unit 100 (motorcycle side) is therefore the wake word engine 306, and subsequent processes and activities are as previously combined with Figure 2 happened as discussed.

[0107] Figure 4 is an example of further development of the embodiments in two directions:

[0108] Multi-user operation, i.e., helmet-controlled operation, may be extended to two or more smart helmets SH, SH2 (for simplicity, only two are given as examples here). For example, helmets worn by different motorcyclists when riding the same motorcycle (such as MB) at different times may have different usage (control) configurations.

[0109] Further increasing the scope / complexity of processing performed at the smart helmet level.

[0110] It should also be understood that along the Figure 4 The two directions of development illustrated in the example can be carried out independently.

[0111] That is, in one or more embodiments:

[0112] like Figure 4 The multi-user operation illustrated in the example can be applied to Figure 2 and Figure 3 The system layout shown in the example,

[0113] Similarly, in Figure 2 and Figure 3 Based on the system layout illustrated in FIG, further expansion of the processing scope / complexity performed at the level of the smart helmet can be achieved as follows Figure 4 , without providing multi-user operation.

[0114] In the latter aspect (i.e. whether or not multi-user operation is considered), Figure 4 In one or more embodiments illustrated in FIG. 1 , in addition to the voice detection processing 302, 304, the wake-up word processing at 306, 308 is also run at the helmet SH and / or SH. This can further reduce battery consumption at the helmet because the communication activity (e.g., Bluetooth) between 200A and 200RX is limited to voice messages that are recognized as containing certain wake-up words.

[0115] This situation can help to enable multi-user activities, as audio streams from different helmets (such as SH and SH2) can be selected in a mutually exclusive manner, for example by considering different wake-up words for different users (multiple users with different wake-up words).

[0116] In such Figure 4 In the embodiment illustrated in FIG. 1 , the first process performed in unit 100 (motorcycle side) is therefore a check of network availability performed at 310, and subsequent processes and activities are as previously described in conjunction with FIG. Figure 2 happened as discussed.

[0117] Figure 5 are examples of implementations that can help obtain faster responses to voice commands.

[0118] In combination with Figure 2 While the system layout illustrated in the example is illustrated, Figure 5 The implementation method can also be applied to Figure 3 and Figure 4 The layout illustrated in (with or without implementing the multi-user option).

[0119] In addition, Figure 5 Components, elements or functions that have been described in connection with previous figures are denoted by like reference numerals, and detailed descriptions of these components, elements or functions are not repeated.

[0120] In brief, in Figure 5 In the embodiment illustrated in , activating local voice processing at 312 and 316 is no longer dependent on the results of the network availability test performed at 310 .

[0121] In such Figure 5 In the embodiment illustrated in , commands are forwarded for processing two paths in parallel, namely the local engines at 312 and 316 and the cloud-based engine C via 314 .

[0122] In such Figure 5 In the embodiment illustrated in , (only) processing on the latter path (e.g., cloud command generator processing at 314) can be conditioned on the network availability test performed at 310, and the system will proceed to EXIT, without performing any processing at 314, and without attempting to transmit to an unavailable network.

[0123] like Figure 5 The operation of the embodiment illustrated in may be based on criteria for selecting a previously received recognition response (i.e., the first received recognition response) (e.g., at node 318) in view of subsequent processing and activities. This may come from a local engine at 312 and 316, or may come from a cloud-based engine C via 314, with the possibility of discarding a subsequent response received from another engine.

[0124] This approach can help obtain faster responses to voice commands.

[0125] For example, in the presence of a command that is (quickly) recognized by the local engine 312, the corresponding action may be performed immediately, without waiting for a (also) possible cloud-based recognition of the same command.

[0126] In the absence of a command recognized by the local engine 312, more sophisticated cloud-based recognition facilities may be relied upon if available.

[0127] In combination with the above embodiments or alternative solutions of some aspects of the above embodiments, Figures 6 to 8 The motorcycle processor 100 is shown connecting a smart device (e.g., smart phone, smart watch, etc.) SD in parallel with the helmet SH. In this case, if the motorcycle processor 100 is not equipped with a remote network enabling system, the smart device SD can provide an Internet connection to the cloud C. The motorcycle processor 100 then acts as a bridge between the helmet SH and the smart device SD.

[0128] like Figure 6As shown in , regarding helmet-motorcycle communication, the helmet SH will "see" the motorcycle processor 100 as a standard smart device and utilize standard communication protocols (BT profiles such as A2DP, HFP, etc.). In order to obtain a continuous audio stream from the helmet SH, the processor 100 can use the Bluetooth hands-free profile (BT HFP) to forward a fake call to the headset that will open a two-way audio stream. The audio stream from the motorcycle to the helmet can be used for both standard purposes (audio calls) and customized purposes (commands, responses, system messages, etc.).

[0129] Regarding smart device-motorcycle communication, the smart device SD can provide an Internet connection to the motorcycle processor 100. The motorcycle is "seen" as a node in the local network.

[0130] like Figure 7 As shown in , in the case of a customized smart helmet CSH, the communication with the motorcycle processor 100 can be enhanced with customized communications capable of delivering dedicated services (e.g., opening an audio stream only for commands, sending accelerometer measurements, etc.).

[0131] like Figure 8 As shown in , again with regard to smart device-motorcycle communication, a smart device SD-DA with dedicated software (e.g., App) can also be integrated with a motorcycle to provide enhanced services. These services may include better local voice recognition systems (smart devices usually have more resources than motorcycle processors), information provision (weather, news, etc.), information collection (data logging, diagnostics, etc.), firmware updates, etc.

[0132] It should be understood that, although advantageous in various aspects, voice activity detection (as illustrated at 302, 304 in the figure) and / or wake-up word detection (as illustrated at 306, 308 in the figure) may not be mandatory features of an embodiment. This applies whether it is implemented on the motorcycle MB (e.g., at the dashboard) or on a "smart" helmet SH, SH2.

[0133] Similarly, while the embodiments as illustrated herein may advantageously implement (on the motorcycle MB and / or in the smart helmet SH, SH2) both voice activity detection (302, 304) and wake-up word detection (306, 308) in order to reduce signal transmission to a command recognition function (locally and / or network-based), one or more embodiments may even provide only one of those functions being implemented in order to reduce signal transmission to the command recognition function.

[0134] As exemplified herein, a dashboard voice control system (e.g., 10) for a motorcycle (e.g., MB) may include:

[0135] A receiver circuit device (e.g., 200RX, 300) is configured to receive a voice generated signal (e.g., provided by MIC, MIC2),

[0136] command recognition circuitry (e.g., 312, 316; 314, 319) configured to recognize a voice-generated command signal for a motorcycle instrument panel from the voice-generated signal received at the receiver circuitry,

[0137] The command implementation circuit device (e.g., 320, 200TX) is configured to implement the action of the motorcycle instrument panel (e.g., D, TT) according to the command signal generated by the voice recognized by the command recognition circuit device,

[0138] The command recognition circuit device comprises:

[0139] a (local) command recognition circuit (e.g., 312) located in a first signal propagation path (e.g., 312, 316, 318) toward the command implementation circuitry, the command recognition circuitry being configured to provide a command signal generated by a voice recognized by the command recognition circuitry to the command implementation circuitry,

[0140] A network connection interface (e.g., 200) located in a second signal propagation path (e.g., 314, 200, 319, 318) toward the command implementation circuit device, the network connection interface being configured to:

[0141] transmitting the voice generated signal to a network-based voice generated command signal recognition device (e.g., C),

[0142] A voice generated command signal recognized by the network-based voice generated command signal recognition device is received from the network-based voice generated command signal recognition device and provided to the command implementation circuitry.

[0143] The motorcycle dashboard voice control system as illustrated herein may include a network availability node (e.g., 310) configured to check the availability of a network-based voice-generated command signal recognition device and suppress transmission of a voice-generated signal toward a network connection interface because the network-based voice-generated command signal recognition device is checked as unavailable (e.g., 310=N).

[0144] A motorcycle dashboard voice control system as exemplified herein may include a network availability node configured to suppress propagation of a voice generated signal toward a command recognition circuit due to a network-based voice generated command signal recognition device being checked as available (e.g., 310=Y).

[0145] As exemplified herein (see Figure 5 ) in a motorcycle dashboard voice control system, a first signal propagation path in a command recognition circuit device can bypass a network availability node (310), wherein regardless of the availability of a network-based voice-generated command signal recognition device, the first signal propagation path (312, 316, 318) is open (i.e., provided or facilitated) to propagate the voice-generated signal toward the command recognition circuit.

[0146] A motorcycle dashboard voice control system as illustrated herein may include a wake-up word processing circuit device (e.g., 306, 308) that is configured to detect the occurrence of at least one wake-up word in voice-generated signals received at a receiver circuit device and (e.g., at 308) suppress forwarding of voice-generated signals to a command recognition circuit device except for those voice-generated signals in which the occurrence of at least one wake-up word is detected.

[0147] A motorcycle dashboard voice control system as exemplified herein may include voice activity detection circuitry (e.g., 302, 304) configured to detect the presence of a voice-generated signal in a signal received at a receiver circuitry and (e.g., at 304) suppress forwarding of signals (e.g., noise) received at the receiver circuitry to the command recognition circuitry, except for signals in which the voice-generated signal presence is detected.

[0148] A motorcycle dashboard voice control system as illustrated herein may include a voice activity detection circuit device arranged upstream of a wake-up word processing circuit device, wherein the voice activity detection circuit device is configured (e.g., at 304) to suppress forwarding of signals received at the receiver circuit device to the wake-up word processing circuit device, except for signals detecting the occurrence of voice-generated signals.

[0149] In the motorcycle dashboard voice control system illustrated in this article, the receiver circuit device (e.g., 200RX) can be configured to receive voice-generated signals from multiple sources (e.g., SH, SH2), and selectively receive voice-generated signals from one source (SH, respectively SH2) of the multiple sources without receiving voice-generated signals from another source (SH2, respectively SH) of the multiple sources.

[0150] A dashboard voice control system as exemplified herein may be used to equip (possibly by retrofitting) a motorcycle as exemplified herein (eg, MB).

[0151] A motorcycle helmet as exemplified herein (e.g., SH, SH2) may include (e.g., see Figures 2 to 6 All images of ):

[0152] a microphone (e.g., MIC) sensitive to voice activity at the helmet, the microphone generating a voice-generated signal based on the voice activity,

[0153] The transmitter circuit (eg, 200A) is configured to transmit a voice-generated signal to a receiver circuit device (eg, 200RX) in a dashboard voice control system as exemplified herein.

[0154] A motorcycle helmet as exemplified herein (see, e.g., Figure 3 ) may include voice activity detection circuitry (e.g., 302, 304) configured to detect the presence of a voice generated signal produced by a microphone in response to voice activity and to suppress transmission from a transmitter circuit in the absence of a voice generated signal produced by the microphone detected by the voice activity detection circuitry.

[0155] A motorcycle helmet as exemplified herein (see, e.g., Figure 4 ) may include a wake-up word processing circuit device (e.g., 306, 308) that is configured to detect the occurrence of at least one wake-up word in speech generated signals generated by the microphone, and to suppress transmission of speech generated signals generated by the microphone from the transmitter circuit except for those speech generated signals generated by the microphone in which the occurrence of the at least one wake-up word is detected.

[0156] In a motorcycle helmet as exemplified herein (see again, for example, Figure 4 ), the voice activity detection circuit device can be arranged upstream of the wake-up word processing circuit device, wherein the voice activity detection circuit device is configured to suppress forwarding signals to the wake-up word processing circuit device in the absence of a voice generated signal produced by a microphone detected by the voice activity detection circuit device.

[0157] The method of controlling a motorcycle instrument panel as exemplified herein may include:

[0158] receiving voice-generated signals (at the motorcycle), including voice-generated command signals for a dashboard of the motorcycle,

[0159] identifying a voice-generated command signal for a motorcycle instrument panel from the received voice-generated signal,

[0160] A command implementation circuit device is provided to implement the action of the motorcycle instrument panel according to the command signal generated by the voice recognized by the command recognition circuit device,

[0161] The command signal generated by the recognized speech may include:

[0162] providing a command recognition circuit located in a first signal propagation path toward the command implementation circuitry, the command recognition circuitry being configured to provide a command signal generated by speech recognized by the command recognition circuitry to the command implementation circuitry,

[0163] providing a network connection interface located in a second signal propagation path toward the command implementation circuitry, and activating the network connection interface to:

[0164] transmitting the voice generated signal to a network-based voice generated command signal recognition device,

[0165] A voice generated command signal recognized by the network-based voice generated command signal recognition device is received from the network-based voice generated command signal recognition device and provided to the command implementation circuitry.

[0166] As exemplified herein (see, e.g., Figure 5 ) may include:

[0167] activating both a first signal propagation path and a second signal propagation path towards a command implementation circuit arrangement, wherein a voice generated command signal is recognized as:

[0168] at only one of the first signal propagation path and the second signal propagation path (e.g., due to unavailability of network / cloud resources or inability of the local recognizer to recognize certain voice-generated commands), or

[0169] at the faster one of the first signal propagation path and the second signal propagation path (e.g., the local identifier 312) earlier than at the slower one of the first signal propagation path and the second signal propagation path (e.g., the cloud-based identifier C), and

[0170] The action of the motorcycle instrument panel is implemented according to the command signal generated by the voice recognized by only one path or the faster one of the first signal propagation path and the second signal propagation path.

[0171] The details and embodiments may vary, by way of example only, with respect to what has been disclosed herein without departing from the scope of protection.

[0172] The scope of protection is determined by the appended claims.

Claims

1. A motorcycle instrument panel voice control system, comprising: a receiver circuit arrangement configured to receive a speech generated signal; a wake-word processing circuit arrangement configured to detect the occurrence of at least one wake-word in the speech generated signal received at the receiver circuit arrangement; command recognition circuitry configured to recognize a voice-generated command signal for a motorcycle instrument panel from among the voice-generated signals received at the receiver circuitry; a command implementation circuit device configured to implement a motorcycle instrument panel action according to a command signal generated by the voice recognized by the command recognition circuit device; The command recognition circuit device comprises: a command recognition circuit located in a first signal propagation path toward the command implementation circuitry, the command recognition circuitry being configured to provide a command signal generated by speech recognized by the command recognition circuitry to the command implementation circuitry; and a network connection interface located in a second signal propagation path toward the command implementation circuit device, the network connection interface being configured to: transmitting the voice generated signal to a network-based voice generated command signal recognition device; receiving, from the network-based voice-generated command signal recognition device, the voice-generated command signal recognized by the network-based voice-generated command signal recognition device; and providing the voice generated command signal to the command implementation circuitry; and A network availability node is configured to check the availability of the network-based voice generated command signal recognition device in response to detecting the occurrence of the at least one wake-up word by the wake-up word processing circuit device.

2. The motorcycle dashboard voice control system according to claim 1, wherein the network availability node is configured as: In response to determining that the network-based voice generated command signal recognition device is checked as unavailable, transmission of the voice generated signal toward the network connection interface is suppressed.

3. The motorcycle dashboard voice control system of claim 2, wherein the network availability node is further configured to: in response to determining that the network-based voice-generated command signal recognition device is checked as available, suppress propagation of the voice-generated signal toward the command recognition circuit.

4. A motorcycle dashboard voice control system as described in claim 2, wherein the first signal propagation path in the command recognition circuit device bypasses the network availability node, and wherein regardless of the availability of the network-based voice-generated command signal recognition device, the first signal propagation path is open to propagate the voice-generated signal toward the command recognition circuit.

5. The motorcycle dashboard voice control system according to claim 1, wherein the wake-up word processing circuit device is further configured as: Except for the speech generated signal in which the occurrence of the at least one wake-up word is detected, forwarding of other speech generated signals towards the command recognition circuitry is suppressed.

6. The motorcycle dashboard voice control system according to claim 5, comprising a voice activity detection circuit device, wherein the voice activity detection circuit device is configured to: detecting an occurrence of the speech generated signal in a signal received at the receiver circuitry; and Signals received at the receiver circuitry are suppressed from being forwarded towards the command recognition circuitry, except for signals in which the occurrence of the speech generated signal is detected.

7. A motorcycle dashboard voice control system according to claim 6, wherein the voice activity detection circuit device is arranged upstream of the wake-up word processing circuit device, and wherein the voice activity detection circuit device is configured to suppress forwarding signals received at the receiver circuit device to the wake-up word processing circuit device except for signals in which the occurrence of the voice generated signal is detected.

8. The motorcycle instrument panel voice control system of claim 1, wherein said receiver circuitry is configured to receive said voice generated signals from a plurality of sources.

9. The motorcycle instrument panel voice control system of claim 8, wherein the receiver circuit device is configured to receive the voice generated signal from one of the plurality of sources without receiving other voice generated signals from another of the plurality of sources.

10. A motorcycle helmet comprising: a microphone sensitive to voice activity at the helmet, the microphone configured to generate a voice-generated signal based on the voice activity; a wake-up word processing circuit arrangement configured to detect the presence of at least one wake-up word in the voice-generated signal; a transmitter circuit configured to transmit the voice-generated signal to a receiver circuit device in a dashboard voice control system; as well as A voice activity detection circuit device is configured to: detecting the presence of the speech-generated signal produced by the microphone; as well as suppressing transmission from the transmitter circuitry in the event that the absence of the voice generated signal produced by the microphone is detected by the voice activity detection circuitry; The voice activity detection circuit device is arranged upstream of the wake-up word processing circuit device, and the voice activity detection circuit device is configured to suppress forwarding signals to the wake-up word processing circuit device when the voice activity detection circuit device detects that there is no signal generated by the voice generated by the microphone.

11. The motorcycle helmet according to claim 10, wherein the wake-up word processing circuit device is further configured to: Suppressing transmission from the transmitter circuit of voice generated signals generated by the microphone other than the voice generated signal generated by the microphone in which the occurrence of at least one wake-up word is detected.

12. A method for controlling a motorcycle instrument panel, the method comprising: receiving, by a receiver circuit arrangement, a voice-generated signal comprising a voice-generated command signal for an instrument panel of the motorcycle; identifying, by command recognition circuitry, a voice-generated command signal for the motorcycle instrument panel from the received voice-generated signal; as well as By means of a command recognition circuit device, a motorcycle instrument panel action is realized according to a command signal generated by the voice recognized by the command recognition circuit device, Recognizing the command signal generated by the voice includes: Check the availability of network-based voice-generated command signal recognition equipment; In response to determining that the network-based voice-generated command signal recognition device is checked as unavailable, providing the voice-generated command signal recognized by the command recognition circuit to the command implementation circuit arrangement via a command recognition circuit, the command recognition circuit being located in a first signal propagation path toward the command implementation circuit arrangement; In response to determining that the network-based voice generated command signal recognition device is checked as available, transmitting the voice generated signal to the network-based voice generated command signal recognition device through a network connection interface, the network connection interface being located in a second signal propagation path toward the command implementation circuit device; and After the transmission, receiving from the network-based voice-generated command signal recognition device via the network connection interface; and The command implementation circuit device is provided with a voice generated command signal recognized by the network-based voice generated command signal recognition device through the network connection interface.

13. The method according to claim 12, comprising: activating both the first signal propagation path and the second signal propagation path toward the command-implementing circuitry; as well as The motorcycle dashboard action is implemented according to the command signal generated by the recognized voice.

14. The method of claim 13, wherein the voice-generated command signal is recognized at only one of the first signal propagation path and the second signal propagation path.

15. The method of claim 13, wherein the voice-generated command signal is recognized earlier at a faster one of the first signal propagation path and the second signal propagation path than at a slower one of the first signal propagation path and the second signal propagation path.

16. The method according to claim 12, comprising: checking availability of the network-based voice-generated command signal recognition device; as well as As a result of the network-based voice generated command signal recognition device being checked as unavailable, transmission of the voice generated signal toward the network connection interface is suppressed.

17. The method according to claim 12, comprising: checking availability of the network-based voice-generated command signal recognition device; as well as As a result of the network-based voice generated command signal recognition device being checked as available, propagation of the voice generated signal toward the command recognition circuit is suppressed.

18. The method according to claim 12, comprising: Detecting, by a wake-up word processing circuit device, the presence of at least one wake-up word in the received voice-generated signal; as well as By means of the wake-up word processing circuitry, forwarding of other speech generated signals towards the command recognition circuitry is suppressed, except for the speech generated signal in which the occurrence of the at least one wake-up word is detected.

19. The method according to claim 18, comprising: detecting, by voice activity detection circuitry, the presence of a received voice generated signal in a signal received by said receiver circuitry; as well as By means of the voice activity detection circuitry, forwarding of signals received at the receiver circuitry, other than signals in which the occurrence of the voice generated signal is detected, is suppressed towards the command recognition circuitry.

Images