Docking station multi-host switching method and system

Abstract

The application relates to the technical field of docking stations, and discloses a docking station multi-host switching method and system. The method comprises the following steps: collecting audio data streams and detecting valid voice frame sequences through a built-in microphone of the docking station, generating a switching instruction number according to the valid voice frame sequences; generating a switching trigger signal based on the switching instruction number, and outputting the switching trigger signal to an interrupt input end of a master control chip to trigger the master control chip to execute a switching operation; in response to the switching operation, a master control chip interrupt service routine reads a current active host number and calculates a target host number, writes a channel selection control signal corresponding to the target host number into a USB signal multiplexing chip, and completes physical switching of a USB-C uplink signal path. The application completely replaces manual physical operation with a voice instruction without modifying any existing switching logic, and realizes hardware-level transparent switching without the support of host software and a network.

Description

Technical Field

[0001] This invention relates to the field of docking station technology, and in particular to a method and system for switching multiple hosts in a docking station. Background Technology

[0002] With the increasing prevalence of multi-host office scenarios, users need to frequently switch between host devices connected to the docking station during workflows such as conference presentations and remote collaboration. Existing multi-host docking stations generally rely on physical buttons or DIP switches as the sole switching trigger method, requiring users to manually touch the device each time they switch, causing significant operational interruptions in continuous work scenarios.

[0003] To address the inconvenience of manual operation, remote switching solutions relying on host software or cloud-based voice assistants have emerged in the market. However, these solutions require the host operating system to be running normally, and the switching command must be transmitted through the host's network interface to the cloud recognition server before returning a control signal. This results in high response latency and strong dependence on network connectivity, rendering them completely ineffective when the host is in BIOS / UEFI mode, in hibernation, or without complete driver installation. Furthermore, there is a risk of privacy leaks due to voice data transmission over the network. Existing technology cannot perform voice recognition locally on the docking station and directly convert the recognition results into hardware trigger signals electrically equivalent to physical buttons. Therefore, it is impossible to achieve hardware-level, driverless, transparent switching without relying on host software or a network. Summary of the Invention

[0004] The main objective of this invention is to provide a method and system for switching multiple hosts in a docking station. Without modifying any existing switching logic, this invention completely replaces manual physical operations with voice commands, achieving hardware-level transparent switching that requires no host software or network support.

[0005] To achieve the above objectives, the present invention provides a method for switching between multiple hosts in a docking station, comprising the following steps: The audio data stream is acquired by the built-in microphone of the expansion dock and the valid voice frame sequence is detected. A switching instruction number is generated based on the valid voice frame sequence. A switching trigger signal is generated based on the switching instruction number, and the switching trigger signal is output to the interrupt input terminal of the main control chip to trigger the main control chip to perform a switching operation; In response to the switching operation, the main control chip interrupt service routine reads the currently active host number and calculates the target host number, writes the channel selection control signal corresponding to the target host number to the USB signal multiplexing chip, and completes the physical switching of the USB-C uplink signal path.

[0006] Optionally, in a first implementation of the first aspect of the present invention, the step of acquiring audio data streams and detecting valid voice frame sequences through the built-in microphone of the expansion dock, and generating a switching instruction number based on the valid voice frame sequence, includes: The voice sound pressure signal is acquired by the built-in microphone of the expansion dock, and the voice sound pressure signal is converted from analog to digital to obtain an audio data stream; The audio data stream is subjected to voice activity endpoint detection to obtain a valid voice frame sequence; The valid voice frame sequence is input into the offline voice recognition chip inside the expansion dock for local offline keyword matching to obtain the switching command number.

[0007] Optionally, in a second implementation of the first aspect of the present invention, the step of inputting the effective voice frame sequence into the offline speech recognition chip inside the expansion dock for local offline keyword matching to obtain a switching instruction number includes: The valid speech frame sequence is subjected to speech content recognition to obtain the recognized text; The recognized text is matched one by one with each instruction word in the instruction word table pre-stored in the offline speech recognition chip. The instruction word that matches completely is taken as the target instruction word, and the switching instruction number corresponding to the target instruction word is output.

[0008] Optionally, in a third implementation of the first aspect of the present invention, when no instruction word in the instruction word list completely matches the recognized text, no number is output and the system is reset to the voice activity endpoint detection waiting state.

[0009] Optionally, in a fourth implementation of the first aspect of the present invention, the step of generating a switching trigger signal based on the switching instruction number and outputting the switching trigger signal to the interrupt input terminal of the main control chip to trigger the main control chip to perform a switching operation includes: Based on the switching instruction number, the signal output terminal of the offline speech recognition chip is switched to an effective level drive state and maintained at a preset pulse width before returning to the default state to obtain a switching trigger signal. The switching trigger signal is connected to the interrupt input terminal of the main control chip, and the interrupt input terminal generates an interrupt signal to trigger the main control chip to perform a switching operation.

[0010] Optionally, in a fifth implementation of the first aspect of the present invention, the step of switching the signal output terminal of the offline speech recognition chip to an effective level drive state based on the switching instruction number and maintaining a preset pulse width before restoring it to the default state to obtain a switching trigger signal includes: Based on the switching instruction number, the offline speech recognition chip signal output terminal is triggered to switch from the default state to the effective level drive state, and the duration is compared with the preset pulse width. When the duration reaches the preset pulse width, the signal output terminal is restored to the default state, and a switching trigger signal is obtained.

[0011] Optionally, in a sixth implementation of the first aspect of the present invention, the docking station multi-host switching method further includes: The debounce output signal generated by the physical button and the low-level active pulse signal are respectively connected to the same convergence node in an open-drain manner. The convergence node is connected to the system power supply through a common pull-up resistor and is also connected to the external interrupt pin of the main control chip. When either the low-level active pulse signal or the debouncing output signal pulls the convergence node potential low, the external interrupt pin detects the falling edge and triggers an interrupt, outputting a switching trigger signal.

[0012] Optionally, in the seventh implementation of the first aspect of the present invention, in response to the switching operation, the main control chip interrupt service routine reads the currently active host number and calculates the target host number, writes a channel selection control signal corresponding to the target host number to the USB signal multiplexing chip, and completes the physical switching of the USB-C uplink signal path, including: In response to the switching operation, the main control chip interrupt service routine is triggered to read the switching lock flag and the currently active host number from the status register; When the switch lock flag is in a locked state, the interrupt service routine returns and drives the status indicator to flash; when the switch lock flag is in a switch-allowed state, the target host number is calculated based on the currently active host number and written to the status register. Write the corresponding channel selection control signal according to the target host number, and switch the ultra-high speed differential signal pair and DisplayPort auxiliary channel in the USB-C uplink signal path to the uplink interface corresponding to the target host number, thus completing the physical switching of the USB-C uplink signal path.

[0013] Optionally, in the eighth implementation of the first aspect of the present invention, the step of writing the corresponding channel selection control signal according to the target host number and switching the ultra-high-speed differential signal pair and DisplayPort auxiliary channel in the USB-C uplink signal path to the uplink interface corresponding to the target host number, thereby completing the physical switching of the USB-C uplink signal path, includes: Based on the target host number, a channel selection control signal is written to the USB signal multiplexing chip; The internal analog switch of the USB signal multiplexing chip is driven to disconnect the ultra-high-speed differential signal pair and the DisplayPort auxiliary channel in the USB-C uplink signal path from the uplink interface corresponding to the current active host number and connect them to the uplink interface corresponding to the target host number. By controlling the switching of the pull-up and pull-down resistor network on the CC line, USB-PD negotiation is initiated to the uplink interface corresponding to the target host number. After the USB-C controller of the target host detects the change in the CC line state, it completes the handshake between the power role and the data role, and completes the physical switching of the USB-C uplink signal path.

[0014] The present invention also provides a docking station multi-host switching system, comprising: The acquisition module is used to acquire audio data streams through the built-in microphone of the expansion dock and detect valid voice frame sequences, and generate switching instruction numbers based on the valid voice frame sequences; The trigger module is used to generate a switching trigger signal based on the switching instruction number, and output the switching trigger signal to the interrupt input terminal of the main control chip to trigger the main control chip to perform a switching operation; The switching module is used to respond to the switching operation. The main control chip interrupt service routine reads the currently active host number and calculates the target host number, writes the channel selection control signal corresponding to the target host number to the USB signal multiplexing chip, and completes the physical switching of the USB-C uplink signal path.

[0015] In summary, this invention constructs a local offline voice recognition path within the expansion dock, converting the recognition result into a switching trigger signal equivalent to a physical button trigger signal. This signal is then converged to the same interrupt input terminal of the main control chip via a convergence circuit. This prevents the main control chip's interrupt service routine from distinguishing the source of the switching signal, thus completely replacing manual physical operation with voice commands without modifying any existing switching logic. Since voice content recognition and command word matching are all completed in a closed loop on the offline voice recognition chip within the expansion dock, the entire recognition and triggering process does not rely on any external network connection or call any interfaces of the host operating system. Switching can be triggered normally even when the host is in BIOS / UEFI interface, hibernation mode, or without installed drivers, achieving complete transparency to the host operating system. Simultaneously, the command word matching mechanism ensures that only valid voice commands trigger switching, preventing false triggers from environmental noise and invalid voice. When the main control chip detects that the switching lock flag is locked, the interrupt service routine returns directly and provides feedback, providing a complete state protection mechanism for the switching process. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the steps of a multi-host switching method for a docking station in one embodiment of the present invention; Figure 2 This is a schematic diagram illustrating the steps for generating a switching instruction number in an embodiment of the present invention; Figure 3 This is a schematic diagram illustrating the steps of outputting the switching trigger signal in an embodiment of the present invention; Figure 4 This is a schematic diagram illustrating the steps of physically switching the USB-C uplink signal path in an embodiment of the present invention; Figure 5 This is a block diagram of the multi-host switching system structure of the expansion dock in an embodiment of the present invention.

[0017] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0019] Reference Figure 1 This embodiment provides a method for switching between multiple hosts in a docking station, including the following steps: S10 collects audio data streams and detects valid voice frame sequences through the built-in microphone of the expansion dock, and generates a switching instruction number based on the valid voice frame sequence. S20: Generate a switching trigger signal based on the switching instruction number, and output the switching trigger signal to the interrupt input terminal of the main control chip to trigger the main control chip to perform the switching operation; S30, in response to the switching operation, the main control chip interrupt service routine reads the currently active host number and calculates the target host number, writes the channel selection control signal corresponding to the target host number to the USB signal multiplexing chip, and completes the physical switching of the USB-C uplink signal path.

[0020] This invention constructs a local offline voice recognition path within the expansion dock, converting the recognition result into a switching trigger signal equivalent to a physical button trigger signal. This signal is then converged to the same interrupt input terminal of the main control chip via a convergence circuit. This makes it impossible for the main control chip's interrupt service routine to distinguish the source of the switching signal, thus completely replacing manual physical operation with voice commands without modifying any existing switching logic. Since voice content recognition and command word matching are all completed in a closed loop on the offline voice recognition chip within the expansion dock, the entire recognition and triggering process does not rely on any external network connection or call any interfaces of the host operating system. Switching can be triggered normally even when the host is in BIOS / UEFI interface, hibernation mode, or without installed drivers, achieving complete transparency to the host operating system. Simultaneously, the command word matching mechanism ensures that only valid voice commands trigger switching, preventing false triggers from environmental noise and invalid voice. When the main control chip detects that the switching lock flag is locked, the interrupt service routine returns directly with feedback, providing a complete state protection mechanism for the switching process.

[0021] In one example, such as Figure 2 The system acquires audio data streams and detects valid speech frame sequences via the dock's built-in microphone, including: S11 acquires voice sound pressure signals through the built-in microphone of the expansion dock and performs analog-to-digital conversion on the voice sound pressure signals to obtain audio data streams; S12, perform speech activity endpoint detection on the audio data stream to obtain a valid speech frame sequence; S13: Input the valid voice frame sequence into the offline voice recognition chip inside the expansion dock for local offline keyword matching to obtain the switching command number.

[0022] In this example, the docking station has a built-in microphone facing the user's operating side. The built-in microphone continuously receives the user's voice sound pressure signal and converts it into an analog voltage signal that varies with the sound pressure amplitude. The analog voltage signal is directly input to the analog-to-digital converter (ADC) input of the offline speech recognition chip inside the docking station. The ADC inside the offline speech recognition chip samples the data at equal intervals according to a preset sampling frequency and preset quantization precision, forming a continuous audio data stream. The above sampling parameters are suitable for the common input specifications of embedded offline speech recognition chips, and no host operating system is required to participate in audio acquisition.

[0023] The voice activity detection and processing unit in the offline speech recognition chip continuously frames the audio data stream according to a preset frame length. The preset frame length can balance the accuracy of voice start and end boundary positioning and the real-time computing load of the chip. After each frame of audio sampling points enters the short-time energy calculation process, the processing unit obtains the short-time energy value of the corresponding frame and writes the short-time energy value into the frame state buffer. The VAD trigger energy threshold is determined during the factory calibration stage. The VAD trigger energy threshold is used to determine whether the voice frame has the energy condition to enter the recognition process. The average short-time energy of the background noise in the silent environment is obtained by collecting no less than 10 seconds of office background noise audio. The average short-time energy of the user's normal pronunciation command is obtained by collecting no less than 20 command pronunciation samples at a distance of 0.5 meters from the microphone. The VAD trigger energy threshold is the arithmetic mean of the above two. After adopting the above threshold setting method, the steady-state air conditioner noise, fan noise and slight ambient noise are lower than the VAD trigger energy threshold. When the user normally says switching commands such as "Host A", "Host B" and "Next", the short-time energy value of the continuous voice frames can stably exceed the VAD trigger energy threshold.

[0024] When the offline speech recognition chip detects that the number of frames with short-term energy values ​​continuously exceeding the VAD trigger energy threshold reaches a preset starting frame number, it marks the first frame that meets the condition as the speech start frame. The duration corresponding to the preset starting frame number can suppress false triggering caused by a single transient impact sound or electrical glitches. With the speech activity flag bit already set, the offline speech recognition chip continues to receive subsequent audio frames and maintains effective acquisition. When the number of frames with short-term energy values ​​continuously falling below the VAD trigger energy threshold reaches a preset ending frame number, it marks the silence boundary after the last valid speech frame as the speech end frame. The silence duration corresponding to the preset ending frame number can cover short pauses in natural speech, avoiding incorrect segmentation of a complete command. The continuous audio frames from the speech start frame to the speech end frame constitute a valid speech frame sequence, which carries status information such as frame number, sampling rate, quantization bit depth, frame length, and speech activity flag bit.

[0025] The offline speech recognition chip sends the valid speech frame sequence into the local offline keyword matching process. This process is completed entirely within the docking station, without transmitting audio data via the uplink USB-C interface or calling any driver interfaces in Windows, Linux, macOS, or BIOS / UEFI environments. The offline speech recognition chip performs keyword determination on the valid speech frame sequence based on a preset instruction dictionary. When the determination result meets the preset matching criteria, it outputs a switching instruction number. This switching instruction number generates a switching trigger signal equivalent to a physical button trigger signal, allowing the main control chip to calculate the target host number based on the currently active host number, following the original physical button switching logic. If the recognition does not meet the preset matching criteria, the offline speech recognition chip does not output a switching instruction number and resets the speech activity detection and processing unit to a waiting state to prevent environmental noise, non-instruction statements, or incomplete speech from entering the subsequent hardware trigger path.

[0026] In one example, S13 includes: Speech content recognition is performed on valid speech frame sequences to obtain the recognized text; The recognized text is matched one by one with each instruction word in the pre-stored instruction word table inside the offline speech recognition chip. The instruction word that matches completely is taken as the target instruction word, and the switching instruction number corresponding to the target instruction word is output.

[0027] In this example, the offline speech recognition chip, having already confirmed the start and end boundaries of speech activity detection, performs speech content recognition processing on the valid speech frame sequence. The processing object no longer includes preceding silent segments, background noise segments, or invalid frames after the speech ends. Therefore, the speech recognition processing unit inside the chip can form a stable recognition input around the pronunciation segment of the complete instruction word. The speech content recognition processing is completed in a closed loop on the offline speech recognition chip inside the docking station, without transmitting audio data via the uplink USB-C interface or calling any driver interfaces of the host operating system. It can be executed normally regardless of the host's operating state.

[0028] After the offline speech recognition chip completes the speech content recognition of a valid speech frame sequence, it obtains the corresponding recognized text. The recognized text is either a word from a finite instruction vocabulary or invalid text that cannot match any word. The offline speech recognition chip pre-stores a finite instruction vocabulary. Each instruction word in the finite instruction vocabulary is bound to a switching instruction number. For example, voice commands such as "Host A", "Host B", and "Next" correspond to different switching instruction numbers. The instruction vocabulary is fixed to the chip's storage area during the product calibration phase and maintains consistency with the host number mapping relationship in the docking station firmware. The chip performs string matching between the recognized text and each instruction in the instruction word list. When the recognized text completely matches an instruction in the instruction word list, that instruction is taken as the target instruction. The switching instruction number bound to the target instruction in the instruction word list is read and output to the signal conversion logic, so that the signal generation circuit can trigger a switching trigger signal equivalent to the physical button trigger signal according to the switching instruction number. When the recognized text fails to completely match any instruction in the instruction word list, the offline speech recognition chip does not output any switching instruction number and resets the speech activity detection and processing unit to the waiting state to prevent environmental noise, non-instruction statements, or partial sentences of speech from entering the subsequent hardware triggering path.

[0029] In one example, when no instruction word in the instruction word list completely matches the recognized text, no number is output and the system is reset to the speech activity endpoint detection waiting state.

[0030] In this example, after the offline speech recognition chip completes the speech content recognition of the valid speech frame sequence, it writes the obtained recognized text into the matching buffer area, and reads each instruction word from the chip's internal instruction word table storage area in turn, and performs string matching between the recognized text and each instruction word one by one.

[0031] The string matching is based on a perfect match. The offline speech recognition chip compares the recognized text with all preset command words in the command word list, such as "host A", "host B", and "next", one by one, and writes each comparison result into the candidate matching register. After all command words have been compared, the offline speech recognition chip searches the candidate matching register to see if there is a command word that perfectly matches the target.

[0032] When a perfectly matched instruction word exists, the offline speech recognition chip takes that instruction word as the target instruction word, reads the switching instruction number bound to the target instruction word in the instruction word table, and sends the switching instruction number to the subsequent switching trigger signal generation logic. When there is no perfectly matched instruction word in the candidate matching register, it means that the recognized text does not match the preset switching instruction word. The offline speech recognition chip does not output any number, clears the current matching cache, clears the candidate matching register, and resets the recognition state to the voice activity endpoint detection waiting state. This prevents environmental noise, similar pronunciations, or non-switching statements from continuing to enter the hardware triggering link, reducing the risk of erroneously triggering host switching.

[0033] In one example, such as Figure 3 A switching trigger signal is generated based on the switching instruction number, and the switching trigger signal is output to the interrupt input terminal of the main control chip to trigger the main control chip to execute the switching operation, including: S21, based on the switching instruction number, the signal output terminal of the offline voice recognition chip is switched to an effective level drive state and maintained at a preset pulse width before returning to the default state, thus obtaining the switching trigger signal; S22 connects the switching trigger signal to the interrupt input terminal of the main control chip. The interrupt input terminal generates an interrupt signal, triggering the main control chip to perform the switching operation.

[0034] In this example, after the offline speech recognition chip outputs a switching command number in the front-end recognition process, the chip's internal output control logic verifies the validity of the switching command number. Only when the switching command number belongs to a host switching number already bound in the preset command vocabulary will the output control logic allow the signal output terminal to enter the pulse generation state. In the default standby state, the signal output terminal maintains its default state and does not interfere with the physical button trigger signal, thus the original trigger path of the physical button can remain independent and effective. After a valid switching command number arrives, the offline speech recognition chip switches the signal output terminal to the valid level drive state, maintains the preset pulse width, and then returns to the default state, thereby forming a switching trigger signal. The preset pulse width is adjusted according to the recognition requirements of the main control chip's interrupt input terminal for the valid level holding time, so that the pulse width of the switching trigger signal is longer than the minimum recognition time of the main control chip and shorter than the minimum valid interval between two adjacent trigger events, thereby reducing the risk of missed detection caused by an excessively narrow switching trigger signal, and reducing the impact of an excessively long continuous valid level time on the response of subsequent trigger events.

[0035] Both the switching trigger signal and the physical button trigger signal are connected to the same aggregation node, which is also connected to the interrupt input of the main control chip. When the offline voice recognition chip does not output a switching trigger signal and the physical button is not triggered, the aggregation node maintains its default level, and the interrupt input of the main control chip is in a stable idle state. When the voice recognition path outputs a switching trigger signal, the potential of the aggregation node changes and an interrupt signal is generated at the interrupt input of the main control chip. When the user presses the physical button, the trigger signal output by the physical button through the debounce circuit also triggers a change in the potential of the aggregation node and generates an interrupt signal at the same interrupt input. Since the two signals are consistent in terms of effective level amplitude, effective polarity, and trigger edge, the interrupt service routine of the main control chip only receives a unified switching trigger signal and does not need to identify whether the switching trigger signal comes from the voice recognition chip or the physical button. Through the above connection method, the voice trigger path is converted into a hardware input equivalent to the physical button trigger path. The original main control chip switching logic, lock judgment logic, and USB signal multiplexing chip driver logic do not need to be modified. The docking station can stably convert voice commands into switching trigger signals that the main control chip can recognize without relying on the host operating system or network connection.

[0036] In one example, S21 includes: The offline speech recognition chip's signal output is triggered by switching the command number to switch from the default state to the effective level drive state, and the duration is compared with the preset pulse width. When the duration reaches the preset pulse width, the signal output is restored to the default state, and a switching trigger signal is obtained.

[0037] In this example, the offline speech recognition chip reads the switching instruction number in its internal output control logic and confirms whether the switching instruction number belongs to the registered host switching instruction range. After confirmation, the output control logic sends a timing start signal to the internal timing unit and switches the signal output terminal from the default state to the valid level drive state, causing the convergence node potential to change. The interrupt input terminal of the main control chip can detect the valid trigger edge. At the same time, the timing unit starts counting incrementally with the chip's internal reference clock as the counting source and forms a timing duration corresponding to the valid level holding time in the counting register.

[0038] The timing unit compares the current timing duration with the preset pulse width in each counting cycle. When the comparison result does not reach the preset pulse width, the signal output terminal maintains a valid driving level, and the aggregation node maintains a valid level, ensuring that the interrupt input terminal of the main control chip obtains a sufficiently stable valid level holding time. When the comparison result reaches the preset pulse width, the output control logic restores the signal output terminal to the default state, and the aggregation node is restored to the default level through a pull-up resistor. The interrupt input terminal completes a complete trigger and recovery process. The above processing ensures that the offline voice recognition chip only outputs a switching trigger signal of the preset pulse width after the valid switching command number arrives, and automatically exits the driving state after the switching trigger signal ends, without interfering with subsequent physical button trigger events or voice trigger events.

[0039] In one example, the docking station multi-host switching method also includes: The debounce output signal and the low-level active pulse signal generated by the physical button are connected to the same aggregation node in open-drain mode. The aggregation node is connected to the system power supply through a common pull-up resistor and is also connected to the external interrupt pin of the main control chip. When either the low-level active pulse signal or the debouncing output signal pulls the convergence node potential low, the external interrupt pin detects the falling edge and triggers an interrupt, outputting a switching trigger signal.

[0040] In this example, when the user presses the physical button, the mechanical contacts will generate multiple short jumps in the initial stage of closure. The debounce circuit absorbs the contact jitter components, preventing the short-term contact jitter from being directly transmitted to the interrupt input terminal of the main control chip. The parameters of the debounce circuit are tuned according to the duration of physical button contact jitter, the input threshold of the interrupt input terminal of the main control chip, and the allowable key response delay, so that the debounce circuit can suppress the short jumps in the initial stage of mechanical contact closure, while ensuring that the effective trigger signal formed by the user's normal press is transmitted within an acceptable response time. After the physical button is held down, the debounce circuit outputs a stable debounce output signal.

[0041] The switching trigger signal output by the offline voice recognition chip and the physical button debounce output signal are both connected to the same aggregation node, which is also connected to the interrupt input of the main control chip. When neither the voice trigger path nor the physical button trigger path is activated, the aggregation node maintains its default level, and the interrupt input of the main control chip is in an idle waiting state. When the voice trigger path outputs a switching trigger signal, the aggregation node's potential changes and generates a valid trigger edge; when the physical button trigger path outputs a debounce output signal, the aggregation node also generates a valid trigger edge. The interrupt input of the main control chip is configured in valid edge trigger mode, so when any signal triggers the aggregation node, the main control chip enters the interrupt service routine and uniformly identifies this trigger as a switching trigger signal. Since the voice trigger path and the physical button trigger path exhibit the same valid level amplitude, the same valid polarity, and the same trigger edge characteristics at the aggregation node, the main control chip does not need to distinguish the trigger source. The original switching state judgment, lockout protection, and USB signal multiplexing chip control processes can all be used, achieving equivalent aggregation of voice triggering and physical button triggering at the interrupt input level.

[0042] In one example, such as Figure 4 In response to the switching operation, the main control chip interrupt service routine reads the currently active host number and calculates the target host number, writes the channel selection control signal corresponding to the target host number to the USB signal multiplexing chip, and completes the physical switching of the USB-C uplink signal path, including: S31. Respond to the switching operation, trigger the main control chip interrupt service routine, and read the switching lock flag and the currently active host number in the status register; S32. When the switch lock flag is in the locked state, the interrupt service routine returns and drives the status indicator to flash; when the switch lock flag is in the switch-allowed state, the target host number is calculated based on the currently active host number and written to the status register. S33. Write the corresponding channel selection control signal according to the target host number, and switch the ultra-high speed differential signal pair and DisplayPort auxiliary channel in the USB-C uplink signal path to the uplink interface corresponding to the target host number, thus completing the physical switching of the USB-C uplink signal path.

[0043] In this example, the interrupt input of the main control chip enters the interrupt service routine after receiving the switching trigger signal. The switching trigger signal can come from the switching trigger signal output by the voice recognition chip or from the trigger signal after physical button debouncing. However, the two types of trigger signals have formed a unified hardware input at the convergence node, so the interrupt service routine only processes the unified switching event. After entering the interrupt service routine, the main control chip reads the switching lock flag and the currently active host number in the status register. The switching lock flag is used to indicate whether the current state is in a switching-allowed state. Using a single-bit status bit can reduce the judgment overhead in the interrupt service routine and facilitate access by the main loop, protection logic, and indicator logic. The currently active host number is used in the dual-host expansion dock to identify the host currently using the uplink signal path. The number is consistent with the channel selection table of the USB signal multiplexing chip, so that subsequent channel control signals can be directly mapped from the host number.

[0044] After the status is read, the main control chip first checks the switch lock flag. When the switch lock flag is locked, it indicates that the docking station is currently in a switch-prohibited cycle. The interrupt service routine does not update the currently active host number, nor does it write a new channel selection control signal to the USB signal multiplexing chip. Instead, it drives the status indicator to flash in a suggestive manner before returning, making it clear to the user that the switch request has not been executed. When the switch lock flag is in a switch-allowed state, the interrupt service routine calculates the target host number based on the currently active host number. In a dual-host scenario, the target host number is obtained by alternating between the two host numbers. After the target host number is determined, the main control chip writes the target host number into the active host number field in the status register and simultaneously sets the switch execution flag, keeping the interrupt service routine's execution time short and avoiding prolonged channel switching and protocol renegotiation operations in the interrupt context.

[0045] The main control chip reads the switching execution flag bit in its main loop or switching task, queries the channel selection control table according to the target host number, and writes the channel selection control signal corresponding to the target host number to the USB signal multiplexing chip. After receiving the channel selection control signal, the USB signal multiplexing chip disconnects the ultra-high-speed differential signal pair and the DisplayPort auxiliary channel in the USB-C uplink signal path from the current uplink interface and connects them to the uplink interface corresponding to the target host number, thus completing the physical switching of the USB-C uplink signal path.

[0046] In one example, S33 includes: Write the channel selection control signal to the USB signal multiplexing chip based on the target host number; The internal analog switch of the USB signal multiplexing chip disconnects the ultra-high-speed differential signal pair and the DisplayPort auxiliary channel in the USB-C uplink signal path from the uplink interface corresponding to the current active host number and connects them to the uplink interface corresponding to the target host number. By controlling the switching of the pull-up and pull-down resistor network on the CC line, USB-PD negotiation is initiated to the uplink interface corresponding to the target host number. After the target host's USB-C controller detects the change in the CC line status, it completes the handshake between the power role and the data role, and completes the physical switching of the USB-C uplink signal path.

[0047] In this example, when performing a physical switch of the USB-C uplink signal path based on the target host number, after the switch execution flag is valid, the main control chip reads the target host number already written in the status register and queries the internal channel selection control table based on the target host number to obtain the channel selection control signal corresponding to the target uplink interface. The main control chip writes the channel selection control signal into the channel selection register of the USB signal multiplexer chip through the control bus, causing the USB signal multiplexer chip to enter the channel reconfiguration state. After receiving the channel selection control signal, the USB signal multiplexer chip first closes the high-speed analog switch path between the uplink interface corresponding to the currently active host number and the docking station's downlink port, causing the ultra-high-speed differential signal pair and the DisplayPort auxiliary channel in the USB-C uplink signal path to disconnect from the current uplink interface. Then, it opens the analog switch path of the uplink interface corresponding to the target host number, connecting the ultra-high-speed differential signal pair and the DisplayPort auxiliary channel to the target uplink interface. The on / off action of the analog switch is an internal hardware-level path reconnection within the chip, and the switching completion time is usually on the order of nanoseconds. This time only represents the change in the conduction relationship of the analog switch and does not include the time required for subsequent protocol negotiation.

[0048] After the differential signal path and DisplayPort auxiliary channel are established, the main control chip continues to control the pull-up and pull-down resistor network corresponding to the CC line, causing the CC line state of the target uplink interface to change in accordance with the USB Type-C connection detection rules. Upon detecting the CC line state change, the target host's USB-C controller initiates the USB-PD negotiation process, confirming the power role, data role, and available power supply capabilities during the negotiation. When the target host supports DisplayPort alternative mode, the video auxiliary channel link is established after the USB-PD negotiation ends, enabling the target host to take over display output and downlink peripheral access. Throughout this process, the main control chip does not need to send control commands through the host operating system, nor does it rely on the host-side driver. The target host perceives a standard USB-C device connection change, while the docking station internally completes the physical path switch from the currently active host's corresponding uplink interface to the target host's corresponding uplink interface, thus completing the hardware-level switch of the USB-C uplink signal path.

[0049] Reference Figure 5 This embodiment provides a docking station multi-host switching system, including: Acquisition module 1 is used to acquire audio data streams through the built-in microphone of the expansion dock and detect valid voice frame sequences, and generate switching instruction numbers based on the valid voice frame sequences; Trigger module 2 is used to generate a switching trigger signal based on the switching instruction number and output the switching trigger signal to the interrupt input terminal of the main control chip to trigger the main control chip to perform the switching operation; Switching module 3 is used to respond to switching operations. The main control chip interrupt service routine reads the currently active host number and calculates the target host number, writes the channel selection control signal corresponding to the target host number to the USB signal multiplexing chip, and completes the physical switching of the USB-C uplink signal path.

[0050] In this embodiment, the specific implementation of each unit in the above system embodiment is described in the above method embodiment, and will not be repeated here.

[0051] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, system, article, or method that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, system, article, or method. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, system, article, or method that includes that element.

[0052] The above description is only a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for switching between multiple hosts in a docking station, characterized in that, include: The audio data stream is acquired by the built-in microphone of the expansion dock and the valid voice frame sequence is detected. A switching instruction number is generated based on the valid voice frame sequence. A switching trigger signal is generated based on the switching instruction number, and the switching trigger signal is output to the interrupt input terminal of the main control chip to trigger the main control chip to perform a switching operation; In response to the switching operation, the main control chip interrupt service routine reads the currently active host number and calculates the target host number, writes the channel selection control signal corresponding to the target host number to the USB signal multiplexing chip, and completes the physical switching of the USB-C uplink signal path.

2. The docking station multi-host switching method according to claim 1, characterized in that, The process of acquiring audio data streams and detecting valid voice frame sequences via the built-in microphone of the expansion dock, and generating switching instruction numbers based on the valid voice frame sequences, includes: The voice sound pressure signal is acquired by the built-in microphone of the expansion dock, and the voice sound pressure signal is converted from analog to digital to obtain an audio data stream; The audio data stream is subjected to voice activity endpoint detection to obtain a valid voice frame sequence; The valid voice frame sequence is input into the offline voice recognition chip inside the expansion dock for local offline keyword matching to obtain the switching command number.

3. The docking station multi-host switching method according to claim 2, characterized in that, The step of inputting the effective voice frame sequence into the offline speech recognition chip inside the expansion dock for local offline keyword matching to obtain the switching command number includes: The valid speech frame sequence is subjected to speech content recognition to obtain the recognized text; The recognized text is matched one by one with each instruction word in the instruction word table pre-stored in the offline speech recognition chip. The instruction word that matches completely is taken as the target instruction word, and the switching instruction number corresponding to the target instruction word is output.

4. The docking station multi-host switching method according to claim 1, characterized in that, When no instruction word in the instruction word list completely matches the recognized text, no number is output and the system is reset to the voice activity endpoint detection waiting state.

5. The docking station multi-host switching method according to claim 4, characterized in that, The step of generating a switching trigger signal based on the switching instruction number and outputting the switching trigger signal to the interrupt input terminal of the main control chip to trigger the main control chip to execute the switching operation includes: Based on the switching instruction number, the signal output terminal of the offline speech recognition chip is switched to an effective level drive state and maintained at a preset pulse width before returning to the default state to obtain a switching trigger signal. The switching trigger signal is connected to the interrupt input terminal of the main control chip, and the interrupt input terminal generates an interrupt signal to trigger the main control chip to perform a switching operation.

6. The docking station multi-host switching method according to claim 5, characterized in that, The step of switching the signal output of the offline speech recognition chip to an effective level drive state based on the switching instruction number, maintaining a preset pulse width, and then restoring it to the default state, yields a switching trigger signal, including: Based on the switching instruction number, the offline speech recognition chip signal output terminal is triggered to switch from the default state to the effective level drive state, and the duration is compared with the preset pulse width. When the duration reaches the preset pulse width, the signal output terminal is restored to the default state, and a switching trigger signal is obtained.

7. The docking station multi-host switching method according to claim 6, characterized in that, The docking station multi-host switching method also includes: The debounce output signal generated by the physical button and the low-level active pulse signal are respectively connected to the same convergence node in an open-drain manner. The convergence node is connected to the system power supply through a common pull-up resistor and is also connected to the external interrupt pin of the main control chip. When either the low-level active pulse signal or the debouncing output signal pulls the convergence node potential low, the external interrupt pin detects the falling edge and triggers an interrupt, outputting a switching trigger signal.

8. The docking station multi-host switching method according to claim 1, characterized in that, In response to the switching operation, the main control chip interrupt service routine reads the currently active host number and calculates the target host number, writes the channel selection control signal corresponding to the target host number to the USB signal multiplexing chip, and completes the physical switching of the USB-C uplink signal path, including: In response to the switching operation, the main control chip interrupt service routine is triggered to read the switching lock flag and the currently active host number from the status register; When the switch lock flag is in a locked state, the interrupt service routine returns and drives the status indicator to flash; when the switch lock flag is in a switch-allowed state, the target host number is calculated based on the currently active host number and written to the status register. Write the corresponding channel selection control signal according to the target host number, and switch the ultra-high speed differential signal pair and DisplayPort auxiliary channel in the USB-C uplink signal path to the uplink interface corresponding to the target host number, thus completing the physical switching of the USB-C uplink signal path.

9. The docking station multi-host switching method according to claim 8, characterized in that, The step of writing the corresponding channel selection control signal according to the target host number and switching the ultra-high-speed differential signal pair and DisplayPort auxiliary channel in the USB-C uplink signal path to the uplink interface corresponding to the target host number, thereby completing the physical switching of the USB-C uplink signal path, includes: Based on the target host number, a channel selection control signal is written to the USB signal multiplexing chip; The internal analog switch of the USB signal multiplexing chip is driven to disconnect the ultra-high-speed differential signal pair and the DisplayPort auxiliary channel in the USB-C uplink signal path from the uplink interface corresponding to the current active host number and connect them to the uplink interface corresponding to the target host number. By controlling the switching of the pull-up and pull-down resistor network on the CC line, USB-PD negotiation is initiated to the uplink interface corresponding to the target host number. After the USB-C controller of the target host detects the change in the CC line state, it completes the handshake between the power role and the data role, and completes the physical switching of the USB-C uplink signal path.

10. A docking station multi-host switching system, characterized in that, The steps for implementing the docking station multi-host switching method according to any one of claims 1 to 9 include: The acquisition module is used to acquire audio data streams through the built-in microphone of the expansion dock and detect valid voice frame sequences, and generate switching instruction numbers based on the valid voice frame sequences; The trigger module is used to generate a switching trigger signal based on the switching instruction number, and output the switching trigger signal to the interrupt input terminal of the main control chip to trigger the main control chip to perform a switching operation; The switching module is used to respond to the switching operation. The main control chip interrupt service routine reads the currently active host number and calculates the target host number, writes the channel selection control signal corresponding to the target host number to the USB signal multiplexing chip, and completes the physical switching of the USB-C uplink signal path.

Images