Laparoscope operation method and system based on voice signals

Abstract

The invention relates to the technical field of voice recognition, in particular to a laparoscope operation method based on voice signals. The method comprises the following steps: acquiring an audio signal; denoising and decomposing the audio signal into a plurality of audio segments; determining a time domain feature based on the amplitude in the audio clip; determining frequency domain characteristics based on the frequency amplitude and quantity of the audio signal frequency domain; color features and intensity features are determined based on the audio clip conversion two-dimensional diagram, and an identity coefficient is calculated based on the four features; calculating an identity code by combining the similarity of the audio clips, and determining a target audio clip; and converting the audio signal of the target audio clip into a text instruction to carry out mirror operation. According to the invention, the accuracy and reliability of mirror operation are improved.

Description

Technical Field

[0001] The present application relates to the field of speech recognition technology, and in particular to a laparoscopic mirror movement method and system based on speech signals. Background Art

[0002] Laparoscopic surgery is a representative technique in minimally invasive surgery. Compared to traditional laparotomy, it offers advantages such as smaller incisions, less pain, and faster recovery, making it highly popular with patients. Laparoscopic procedures can be performed through the abdominal or pelvic cavity, reaching hidden lesions that are difficult to detect with traditional imaging. The surgeon observes the lesion under the microscope and then treats the affected area. Currently, minimally invasive laparoscopic surgery is often performed by assistant surgeons following the surgeon's instructions.

[0003] Due to the complexity of the patient's condition, the long surgical process can fatigue the assistant physicians, leading to camera deviation and shaking during the operation. If the patient's condition is complex, multiple assistant physicians may be required, and the degree of coordination between the surgeons can significantly impact the success rate of the operation. Furthermore, unclear speech recognition during the operation can significantly impact the success rate, thus necessitating a precise speech recognition method. Summary of the Invention

[0004] In order to solve the technical problem of unclear voice recognition, this application provides a laparoscopic mirror operation method and system based on voice signals. The technical solutions adopted are as follows: In a first aspect, the present application proposes a laparoscopic operation method based on a voice signal, the method comprising the following steps: Collect audio signals; Decomposing and denoising the audio signal; dividing the denoised audio signal into multiple audio segments using a preset threshold; The instability coefficient is obtained based on the mean of the amplitude in the audio clip, and the time domain characteristics of the audio clip are calculated based on the instability coefficient of the audio clip and the amplitude of the overall signal; the audio signal is converted into the frequency domain and recorded as a frequency domain audio clip; the frequency domain characteristics of the audio clip are obtained based on the number, frequency value and amplitude of the frequencies in the frequency domain audio clip; the audio clip is converted into a two-dimensional image, and the intensity characteristics and color characteristics are determined based on the number of intersections and color entropy in the two-dimensional image; the identity coefficient is calculated based on the time domain characteristics, frequency domain characteristics, intensity characteristics and color characteristics; a key audio clip is preset, and an identity code is calculated based on the similarity between the remaining audio clips and the key audio clip and the identity coefficient of the audio clip, and the identity code is compared with a preset threshold to determine the target audio clip; Convert the audio signal of the target audio clip into text instructions for camera movement.

[0005] In the above solution, the present application analyzes the time-domain variation characteristics of the doctor's oral instruction audio during the camera movement operation, and calculates and characterizes the tone and emotional state of the doctor's oral instruction during the operation by calculating the sound intensity energy; at the same time, the present invention application converts the time-domain signal into the frequency domain, and calculates the timbre center of gravity of the oral instruction audio to characterize the bright and deep voice characteristics of the doctor performing the camera movement operation. Further, in order to accurately obtain the identity of the speaker during the operation, the present application maps the one-dimensional audio signal into a two-dimensional image signal. It more fully and comprehensively reflects the voice change characteristics of the surgeon during the laparoscopic operation. At the same time, in order to ensure the continuity of the behavior change during the camera movement control process, the instructions are spliced, thereby avoiding the phenomenon of poor accuracy caused by insufficient data samples faced by the neural network model during training, and improving the accuracy and reliability of the camera movement operation.

[0006] In one embodiment, the method of dividing the denoised audio signal into multiple audio segments with a preset threshold is as follows: Take a preset multiple of the maximum value of the amplitude in the denoised audio signal as the silence threshold; Record the operation audio signal with an amplitude greater than or equal to the silence threshold as the voice signal, and record the operation audio signal with an amplitude less than the silence threshold as the silence signal, where the operation audio signal is the audio signal after removing noise interference; In the voice signal, each continuously time-connected part is taken as a segment and recorded as an audio segment.

[0007] In one embodiment, the instability coefficient is the number of moments in the audio segment where the amplitude is greater than the average value of all amplitudes in the audio segment.

[0008] In one embodiment, the time-domain feature is negatively correlated with the instability coefficient and positively correlated with the overall amplitude of the signal.

[0009] In one embodiment, the method of obtaining the frequency-domain features of the audio segment based on the number of frequencies, frequency values, and amplitudes of frequencies in the frequency-domain audio segment is as follows: , represents the frequency value of the c-th frequency in the frequency-domain audio segment, represents the amplitude value of the c-th frequency in the frequency-domain audio segment, represents the number of frequencies in the frequency-domain audio segment corresponding to the i-th audio segment, represents the timbre center of gravity of the i-th audio segment, and the timbre center of gravity is a frequency-domain feature.

[0010] In one embodiment, the method of converting the audio segment into a two-dimensional image and determining the intensity feature and color feature based on the number of intersection points and color entropy in the two-dimensional image is as follows: Convert the audio segment into a two-dimensional image corresponding to the audio segment through the Gram angle field; where the peak value of the audio signal corresponds to the intersection point of the two-dimensional image, and record the number of intersection points in the two-dimensional image as the intensity feature; calculate the color entropy of the two-dimensional image as the color feature.

[0011] In one embodiment, the method for calculating the identity coefficient based on the time-domain feature, frequency-domain feature, intensity feature, and color feature is as follows: Take the product of the time-domain feature, frequency-domain feature, intensity feature, and color feature as the identity coefficient.

[0012] In one embodiment, the method for presetting the key audio segment, calculating the identity code based on the similarity between the remaining audio segments and the key audio segment, and comparing the identity code with the preset threshold to determine the target audio segment is as follows: Record the audio segment with the earliest time sequence in the audio segment as the key audio segment; Calculate the identity code based on the DTW distance between the audio segment and the key audio segment and the identity coefficient of the audio segment; The audio segment with the identity code greater than the preset threshold and the key audio segment are from the same speaker, and record the audio segment with the identity code greater than the empirical threshold as the target audio segment.

[0013] In one embodiment, the expression of the identity code is: , represents the identity coefficient of the i-th audio segment, represents the first audio segment, represents the i-th audio segment, represents the DTW distance function, represents the normalization function, represents the identity code of the i-th audio segment.

[0014] In a second aspect, an embodiment of the present application further provides a laparoscopic lens movement system based on a voice signal, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any one of the above-mentioned laparoscopic lens movement methods based on a voice signal.

[0015] The beneficial effects of the present application are as follows: Analyze the time-domain variation characteristics of the doctor's oral command audio during the camera operation of this application, and calculate and characterize the tone and emotional state of the doctor's oral commands during the operation by calculating the sound intensity energy; at the same time, this invention application converts the time-domain signal into the frequency domain, calculates the timbre center of gravity of the oral command audio, and characterizes the bright and deep voice characteristics of the camera operation doctor. Further, in order to accurately obtain the identity of the speaker during the operation, this application maps the one-dimensional audio signal into a two-dimensional image signal. It more fully and comprehensively reflects the voice change characteristics of the surgeon during the laparoscopic operation. At the same time, in order to ensure the continuity of behavior changes during the laparoscopic camera control process, the commands are spliced, thus avoiding the poor accuracy caused by insufficient data samples faced by the neural network model during training, and improving the accuracy and reliability of the laparoscopic camera operation. Brief Description of the Drawings

[0016] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of this application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0017] Figure 1 It is a flowchart of a laparoscopic camera operation method based on voice signals provided by an embodiment of this application. Detailed Embodiments

[0018] To further elaborate on the technical means and effects adopted by this application to achieve the predetermined invention purpose, the following, in combination with the drawings and preferred embodiments, details the specific embodiments, structures, features, and effects of a laparoscopic camera operation method and system based on voice signals proposed according to this application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which this application belongs.

[0020] An embodiment of a laparoscopic camera operation method and system based on voice signals: The following specifically describes the specific solutions of a laparoscopic camera operation method and system based on voice signals provided by this application in combination with the drawings.

[0021] Please refer to Figure 1, which shows a flowchart of a laparoscopic lens operation method based on voice signals provided by an embodiment of the present application. The method includes the following steps: Step S001, collect audio signals.

[0022] During clinical operations, since some lesion areas are relatively hidden, in order to obtain the best viewing position, it is necessary to adjust the lens position of the exploration device to ensure that the exploration device can focus on the lesion area. To better manipulate the movement of the exploration device, an audio collection device can be placed on the physician to collect the audio signals when the physician issues control instructions for the exploration device, and use this audio signal to control the movement of the exploration device. The exploration device is a laparoscope.

[0023] Thus, the audio signals are obtained.

[0024] Step S002, denoise and decompose the audio signals into multiple audio segments.

[0025] During the operation, a relatively large number of monitoring instruments will be placed, and these monitoring instruments will emit periodic beeping sounds. The audio collection device will also record these beeping sounds, resulting in noise in the collected audio signals, which is not conducive to the recognition of audio signals.

[0026] Therefore, in the present application, the audio signals are first used as the input of the VMD variational mode decomposition algorithm to decompose the audio signals into multiple sub-signals. Since there are significant differences in audio characteristics between the beeping sounds of the monitoring instruments and the audio of the physician issuing control instructions. Therefore, the sub-signals obtained after the VMD algorithm decomposition will contain the beeping sounds of the monitoring instruments, and the VMD algorithm can filter out the beeping sounds of the monitoring instruments while maximizing the retention of the audio signals when the physician issues control instructions.

[0027] The audio signals after removing noise interference are denoted as operation audio signals.

[0028] Generally speaking, when the physician issues control instructions, there will be sound compared to the silent operating room, that is, when the physician issues control instructions, the amplitude of the operation audio signal is relatively large, and the operation audio signal at this time is denoted as the voice signal; when no instructions are issued, the amplitude of the operation audio signal is extremely small, and the operation audio signal at this time is denoted as the silent signal; therefore, it is necessary to define the boundary between the two in the operation audio signal and divide the two.

[0029] A certain multiple of the maximum value of the amplitude in the operation audio signal is denoted as the silent threshold, the operation audio signal with an amplitude greater than or equal to the silent threshold is denoted as the voice signal, and the operation audio signal with an amplitude less than the silent threshold is denoted as the silent signal.

[0030] In this embodiment, the maximum value of the amplitude of the operation audio signal is extracted, and the product of the maximum value of the amplitude and a preset parameter is used as the mute threshold, and the preset parameter is set to 0.05.

[0031] Since the operation audio signal when the physician issues a control instruction is relatively concise, the voice signal appears in the form of multiple segments. Each time-continuous part in the voice signal is taken as a segment; thus, the voice signal is divided into multiple audio segments, and each audio segment represents an oral operation instruction. All the divided audio segments are denoted as the audio segment set.

[0032] So far, the audio segment set has been obtained.

[0033] Step S003: Determine the time-domain feature based on the amplitude in the audio segment; determine the frequency-domain feature based on the frequency amplitude and quantity in the frequency domain of the audio signal; determine the color feature and intensity feature based on the two-dimensional graph transformed from the audio segment, calculate the identity coefficient based on the four features; and calculate the identity code in combination with the similarity of the audio segments to determine the target audio segment.

[0034] During the operation of the exploration device, since the situation is relatively complex, the physician may issue multiple operation instructions in a short time when dealing with it; and during the operation, there are a surgeon and an assistant surgeon. The assistant surgeon cooperates with the surgeon to complete the treatment of the lesion. However, for the operation control of the exploration device, it is necessary to distinguish the identities of the assistant surgeon and the surgeon so that the exploration device can more clearly understand the control instructions.

[0035] For each audio segment, calculate the mean value of the amplitudes at all moments in the audio segment, and construct a horizontal line with this mean value. If the number of moments when the amplitude of the audio segment exceeds the mean horizontal line is more, it indicates that the audio signal in this segment fluctuates more significantly. The more obvious the fluctuation, the more unstable the voice. Denote the number of moments when the amplitude exceeds the mean horizontal line as the instability coefficient. In addition, if the overall amplitude of the audio in this audio segment is relatively high, it indicates that the signal is relatively urgent when the instruction is issued. If the audio signal in the audio segment is relatively concentrated and the overall amplitude is relatively low, then the instruction is issued relatively gently.

[0036] Therefore, calculate the sound intensity energy of the audio segment based on the instability coefficient of the audio segment and the overall amplitude of the signal.

[0037] The sound intensity energy is negatively correlated with the instability coefficient and positively correlated with the overall amplitude of the signal.

[0038] It should be noted that positive correlation means that when one variable increases, the other variable also increases, and the change directions of the two variables are the same. When one variable changes from large to small or from small to large, the other variable also changes from large to small or from small to large; the specific relationship is determined by the actual application, and this application does not make special restrictions.

[0039] It should be noted that negative correlation means that when one variable increases, the other variable decreases accordingly. The change directions of the two variables are opposite. When one variable changes from large to small or from small to large, the other variable also changes from small to large or from large to small. The specific relationship is determined by actual applications, and this application does not impose special restrictions.

[0040] Preferably, in this embodiment, the expression of sound intensity energy is: , represents the amplitude at the k-th moment in the audio segment, represents the instability coefficient of the i-th audio segment, represents the number of moments in the i-th audio segment, represents the sound intensity energy of the i-th audio segment. The magnitude of the sound intensity energy reflects the smoothness of the doctor's tone when issuing operation instructions and also represents the time-domain characteristics of the audio.

[0041] The audio segment is a display of the doctor's oral narration in the time domain state, but the audio segment in the time domain state cannot accurately represent the audio characteristics of a specific person. Therefore, the audio segment is used as the input of the FFT algorithm to obtain the frequency-domain audio segment, that is, each frequency-domain audio segment corresponds to an audio segment.

[0042] Based on the number of frequencies, frequency values, and amplitudes of frequencies in the frequency-domain audio segment, the timbre center of gravity of the audio segment is obtained.

[0043] Preferably, the expression of the timbre center of gravity is: , represents the frequency value of the c-th frequency in the frequency-domain audio segment, represents the amplitude value of the c-th frequency in the frequency-domain audio segment, represents the number of frequencies in the frequency-domain audio segment corresponding to the i-th audio segment, represents the timbre center of gravity of the i-th audio segment.

[0044] By calculating the proportion weight of each frequency component and the magnitude of the frequency power, the frequency center of gravity of the frequency-domain signal of the oral instruction audio segment can be obtained. If the speaker's voice is brighter, the value of this frequency center of gravity is larger; conversely, if the speaker's voice is deeper, the calculated value of the frequency center of gravity is smaller. The timbre center of gravity represents the timbre characteristics of the doctor and also represents the frequency-domain characteristics.

[0045] So far, to a certain extent, the audio segments can be distinguished through time-domain characteristics and frequency-domain characteristics, but it is still unknown which doctor the specific audio segment corresponds to, and it is not clear what the control instruction is. Therefore, further analysis is carried out.

[0046] Time-domain and frequency-domain features can only reflect part of the features of the frequency-domain signal of an audio segment. To more comprehensively describe the features of different speaking physicians in the oral instruction audio segment signal, the audio segment and its corresponding identity marker are obtained, and the audio segment is converted into a two-dimensional image corresponding to the audio segment by using the Gram angular field. The identity marker is used to label the audio signal to different physicians.

[0047] In the two-dimensional image, the peak value of the original audio signal will be mapped into an intersection point in the two-dimensional image. Therefore, the present application further calculates and analyzes the two-dimensional image. First, the color entropy of the two-dimensional image is calculated, which represents the color feature of the image and can also represent the color feature of each audio segment. Then, the number of intersection points in the two-dimensional image is counted, and this number represents the intensity feature in the audio signal.

[0048] Therefore, the time-domain feature, frequency-domain feature, color feature, and intensity feature are combined to calculate the identity coefficient of each audio segment.

[0049] Preferably, in this embodiment, the expression of the identity coefficient is: , represents the sound intensity energy of the i-th audio segment, represents the timbre centroid of the i-th audio segment, represents the number of intersection points in the two-dimensional image corresponding to the i-th audio segment, represents the color entropy of the two-dimensional image corresponding to the i-th audio segment, represents the identity coefficient of the i-th audio segment.

[0050] Since every time the probe device is moved, the control instruction is always issued by the surgeon first, the earliest segment in all the collected audio segments is used as the audio segment of the surgeon, that is, the key audio segment.

[0051] Calculate the DTW distance between each audio segment and the key audio segment, and obtain the identity code of the audio segment by combining the identity coefficient of the audio segment.

[0052] Preferably, the expression of the identity code is: , represents the identity coefficient of the i-th audio segment, represents the key audio segment, represents the i-th audio segment, represents the DTW distance function, represents the normalization function, represents the identity code of the i-th audio segment.

[0053] Set an experience threshold. If the identity code is greater than the experience threshold, it indicates that the speaker of this audio segment is the same as that of the first audio segment. Thus, for all audio segments, the audio segments in which the surgeon issues control instructions are determined as target audio segments.

[0054] So far, all the audio segments in which the surgeon issues control instructions have been obtained.

[0055] Step S004: Convert the audio signal of the target audio segment into a text instruction for camera movement operation.

[0056] Use the audio segments in which the surgeon issues control instructions as the input of the LSTM model, so that the exploration device can recognize the text instructions of the surgeon and then perform camera movement operations.

[0057] Since the camera movement operation should have the characteristic of continuous operation, the distance between the camera and the patient's tissue can be roughly divided into long-distance, medium-distance, and close-distance views. Different surgical operation scenarios are applicable to different view distances.

[0058] When the distance is very close, it should move slowly and smoothly to avoid excessive movement. Of course, the distance of the view is constantly changing during the operation process. For example, when using an electrocautery or ultrasonic scalpel for cauterization, in order to prevent the splashing liquid from contaminating the lens, it is necessary to change from medium and close distances to long distance; when the cauterization is completed, it should return to the original view distance in a timely manner; when the surgeon sutures the tissue, the needle should enter and exit at medium or close distance, and when tying a knot, it should retreat to long distance, so as to facilitate the surgeon to observe the positions of the double-handed instruments and the needle comprehensively, which is convenient for operation and avoids damage.

[0059] Therefore, in order to ensure the continuity of camera movement, the text instructions obtained from two audio segments are used to operate the camera movement, so as to minimize or avoid the camera movement dullness phenomenon caused by discontinuous instructions during the voice control process of camera movement.

[0060] So far, the camera movement of the laparoscope has been completed based on speech signal text recognition.

[0061] Based on the same inventive concept as the above method, an embodiment of the present invention also provides a laparoscope camera movement system based on speech signals, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any one of the above methods of a laparoscope camera movement method based on speech signals.

[0062] It should be noted that the above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

[0063] The various embodiments in this specification are all described in a progressive manner. For the same or similar parts between the various embodiments, reference can be made to each other. The key point of each embodiment is to illustrate the differences from other embodiments.

Claims

1. A laparoscopic lens movement method based on voice signals, characterized in that, The method includes the following steps: Collect an audio signal; Decompose and denoise the audio signal; divide the denoised audio signal into multiple audio segments based on a preset threshold; Obtain an instability coefficient based on the mean value of the amplitudes in the audio segment, calculate the time-domain characteristics of the audio segment based on the instability coefficient of the audio segment and the amplitude of the overall signal; convert the audio signal into the frequency domain, denoted as the frequency-domain audio segment; obtain the frequency-domain characteristics of the audio segment based on the number of frequencies, frequency values, and amplitudes of the frequencies in the frequency-domain audio segment; convert the audio segment into a two-dimensional image, and determine the intensity feature and color feature based on the number of intersection points and color entropy in the two-dimensional image; calculate an identity coefficient based on the time-domain characteristics, frequency-domain characteristics, intensity feature, and color feature; preset a key audio segment, calculate an identity code based on the similarity between the remaining audio segments and the key audio segment and the identity coefficient of the audio segment, and compare the identity code with a preset threshold to determine the target audio segment; Convert the audio signal of the target audio segment into a text command for camera movement operation.

2. The laparoscopic lens operation method based on voice signals according to claim 1, characterized in that, The method of dividing the denoised audio signal into multiple audio segments based on a preset threshold is as follows: Take a preset multiple of the maximum value of the amplitudes in the denoised audio signal as the silence threshold; Denote the operation audio signal with an amplitude greater than or equal to the silence threshold as a voice signal, and denote the operation audio signal with an amplitude less than the silence threshold as a silence signal, where the operation audio signal is the audio signal after removing noise interference; In the voice signal, each continuously time-connected part is taken as a segment, denoted as an audio segment.

3. The laparoscopic lens operation method based on voice signals according to claim 1, characterized in that, The instability coefficient is the number of moments when the amplitude in the audio segment is greater than the mean value of all amplitudes in the audio segment.

4. The laparoscopic lens movement method based on voice signals according to claim 1, wherein The time-domain characteristics are negatively correlated with the instability coefficient and positively correlated with the amplitude of the overall signal.

5. The laparoscopic lens manipulation method based on voice signals according to claim 1, wherein The method of obtaining the frequency-domain characteristics of the audio segment based on the number of frequencies, frequency values, and amplitudes of the frequencies in the frequency-domain audio segment is as follows: , represents the frequency value of the c-th frequency in the frequency-domain audio segment, represents the amplitude value of the c-th frequency in the frequency-domain audio segment, represents the number of frequencies in the frequency-domain audio segment corresponding to the i-th audio segment, represents the timbre center of gravity of the i-th audio segment, and the timbre center of gravity is a frequency-domain feature.

6. The laparoscopic lens movement method based on voice signals according to claim 1, characterized in that, The method of converting the audio segment into a two-dimensional image and determining the intensity feature and color feature based on the number of intersection points and color entropy in the two-dimensional image is as follows: Convert the audio segment into a two-dimensional image corresponding to the audio segment through the Gramian Angular Field; where the peak value of the audio signal corresponds to the intersection point of the two-dimensional image, and denote the number of intersection points in the two-dimensional image as the intensity feature; Calculate the color entropy of the two-dimensional image as the color feature.

7. The laparoscopic lens operation method based on voice signals according to claim 1, wherein The method of calculating the identity coefficient based on the time-domain characteristics, frequency-domain characteristics, intensity feature, and color feature is as follows: Take the product of the time-domain characteristics, frequency-domain characteristics, intensity feature, and color feature as the identity coefficient.

8. The laparoscopic lens operation method based on voice signals according to claim 1, characterized in that The method of presetting a key audio segment, calculating an identity code based on the similarity between the remaining audio segments and the key audio segment and the identity coefficient of the audio segment, and comparing the identity code with a preset threshold to determine the target audio segment is as follows: Denote the audio segment with the earliest time sequence in the audio segments as the key audio segment; Calculate the identity code based on the DTW distance between the audio segment and the key audio segment and the identity coefficient of the audio segment; The audio segment with an identity code greater than the preset threshold and the key audio segment are from the same speaker, and denote the audio segment with an identity code greater than the empirical threshold as the target audio segment.

9. The laparoscopic lens-operating method based on voice signals according to claim 8, wherein The expression of the identity code is: , represents the identity coefficient of the i-th audio segment, represents the first audio segment, represents the i-th audio segment, represents the DTW distance function, represents the normalization function, represents the identity code of the i-th audio segment.

10. A laparoscopic lens control system based on voice signals, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of a laparoscopic lens operation method based on voice signals as described in any one of claims 1-9.

Images