Intelligent sound box, control method and device thereof and medium

Abstract

The embodiment of the invention provides an intelligent loudspeaker box, a control method and device thereof and a medium. The intelligent loudspeaker box comprises an audio playing module, a control module and a fragrance blending module. The fragrance blending module comprises a base liquid storage bin, a metering pump, a gas-liquid mixing and atomizing cavity and an atomizing driver; the control module comprises an audio analysis unit and a proportion generation unit; the audio analysis unit is configured to collect an audio signal of the audio playing module and output an emotion label according to the audio signal; the proportion generation unit is configured to generate a fragrance proportion parameter according to the emotion label; determining a target extraction amount of each metering pump according to the fragrance proportioning parameter; each metering pump is controlled to extract the target extraction amount of the fragrance base liquid from the corresponding base liquid storage bin, and the fragrance base liquid is conveyed to the gas-liquid mixing atomization cavity; and the atomization driver is configured to mix and atomize the fragrance base liquid entering the gas-liquid mixing atomization cavity so as to discharge the fragrance base liquid.

Description

Technical Field

[0001] This specification relates to the field of smart speaker technology, and in particular to a smart speaker and its control method, device and medium. Background Technology

[0002] Existing smart speakers have relatively limited functionality, offering only an auditory experience. Traditional fragrance diffusers rely on manual switches or timed spray mechanisms to release scents, resulting in limited fragrance types and requiring manual addition of essential oils to change the aroma. Because smart speakers and fragrance diffusers operate independently and lack intelligent linkage, the diffuser cannot perceive the atmosphere or emotion conveyed by the content played by the smart speaker. This can easily lead to a mismatch between the released fragrance and the emotional content (e.g., releasing an stimulating scent while playing sad music), creating a disconnect between the auditory and olfactory experiences and failing to create a truly immersive sensory environment that integrates both senses.

[0003] Therefore, there is an urgent need for a smart speaker and its control method that can automatically match the playback content with the fragrance function, releasing a fragrance that matches the mood of the playback content, thereby solving the problem of a disconnect between user experience and the music. Summary of the Invention

[0004] This specification provides one or more embodiments of a smart speaker, the smart speaker comprising: an audio playback module, a control module, and a fragrance blending module; the fragrance blending module comprising a base liquid storage chamber, a metering pump, a gas-liquid mixing atomizing chamber, and an atomizing driver; the control module comprising an audio analysis unit and a proportioning generation unit; the audio analysis unit being configured to acquire audio signals from the audio playback module and output emotion tags based on the audio signals; the proportioning generation unit being configured to: generate fragrance proportioning parameters based on the emotion tags; determine the target extraction volume of each metering pump based on the fragrance proportioning parameters; control each metering pump to extract the target extraction volume of fragrance base liquid from the corresponding base liquid storage chamber and deliver the fragrance base liquid to the gas-liquid mixing atomizing chamber; the atomizing driver being configured to mix and atomize the fragrance base liquid entering the gas-liquid mixing atomizing chamber for discharge.

[0005] This specification provides one or more embodiments of a control method for a smart speaker. The control method includes: acquiring audio signals from an audio playback module and outputting emotion tags based on the audio signals; generating fragrance blending parameters based on the emotion tags; determining the target extraction volume of each metering pump based on the fragrance blending parameters; controlling each metering pump to extract the target extraction volume of fragrance base liquid from its corresponding base liquid storage chamber and delivering the fragrance base liquid to a gas-liquid mixing atomization chamber; and controlling an atomization driver to mix and atomize the fragrance base liquid entering the gas-liquid mixing atomization chamber for discharge.

[0006] This specification provides a control device for a smart speaker through one or more embodiments. The device includes at least one processor and at least one memory. The at least one memory is used to store computer instructions. The at least one processor is used to execute at least a portion of the computer instructions to implement the control method of the smart speaker.

[0007] This specification provides one or more embodiments of a computer-readable storage medium that stores computer instructions that, when executed by a processor, implement a control method for a smart speaker. Attached Figure Description

[0008] This specification will be further described by way of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting; in these embodiments, the same reference numerals denote the same structures, wherein:

[0009] Figure 1 These are schematic diagrams illustrating application scenarios of smart speakers based on some embodiments of this specification; Figure 2 This is a structural schematic diagram of a smart speaker according to some embodiments shown in this specification; Figure 3 yes Figure 2 A magnified view of a section at point A in the middle; Figure 4 This is an exemplary flowchart of a control method for a smart speaker according to some embodiments of this specification; Figure 5 This is an exemplary flowchart illustrating the adjustment of the target extraction amount according to some embodiments of this specification; Figure 6 This is a schematic diagram illustrating the fragrance formulation parameters according to some embodiments of this specification.

[0010] Figure label: 10. Smart speaker; 11. Audio playback module; 12. Control module; 121. Audio analysis unit; 1211. Microphone array; 1212. Analysis unit; 122. Proportion generation unit; 123. Proportion correction unit; 124. Preference adaptation unit; 13. Fragrance blending module; 131. Base liquid storage tank; 132. Metering pump; 133. Gas-liquid mixing atomization chamber; 1331. Guide channel; 134. Atomization driver; 14. Partition; 15. Speaker housing; 16. Air duct system; 161. Air duct outlet; 162. Airflow device; 163. Air duct; 1631. First section; 1632. Second section; 1633. Third section; 164. Air duct inlet. Detailed Implementation

[0011] To more clearly illustrate the technical solutions of the embodiments in this specification, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are merely some examples or embodiments of this specification. For those skilled in the art, these drawings can be applied to other similar scenarios without creative effort. Unless obvious from the context or otherwise specified, the same reference numerals in the drawings represent the same structures or operations.

[0012] Figure 1 These are schematic diagrams illustrating application scenarios of smart speakers based on some embodiments of this specification.

[0013] In some embodiments, smart speakers can automatically match the playback content with their own fragrance function, releasing a fragrance that matches the mood of the playback content. At the same time, smart speakers have a wide range of applications and can be used in scenarios such as home living environments, commercial and leisure spaces, hotel rooms, and personal office areas.

[0014] In some embodiments, such as Figure 1 As shown, the application scenario of the smart speaker (hereinafter referred to as application scenario 100) may include the smart speaker 10, the user terminal 120, the network 130, the processor 140, and the database 150.

[0015] The smart speaker 10 can process and / or output audio-related data and fragrance-related data. In some embodiments, the smart speaker 10 may include devices such as a speaker and a digital media player.

[0016] User terminal 120 is a terminal device that interacts with a user. For example, a user terminal may include a mobile phone 120-1, a tablet 120-2, a computer 120-3, etc. A user refers to one or more users using the smart speaker 10.

[0017] Network 130 may include any suitable network capable of facilitating information and / or data exchange. In some embodiments, at least one component of application scenario 100 (e.g., smart speaker 10, user terminal 120, processor 140, database 150, etc.) may exchange information and / or data with at least one other component in application scenario 100 via network 130. For example, processor 140 may obtain relevant information about smart speaker 10, user terminal 120, and database 150 via network 130.

[0018] In some embodiments, network 130 can be any one or more of wired or wireless networks. For example, network 130 may include cable networks, fiber optic networks, telecommunications networks, cable connections, or any combination thereof. Network connections between components may employ one or more of the above methods. In some embodiments, the network may be a point-to-point, shared, centralized, or other topologies, or a combination of multiple topologies. In some embodiments, network 130 may include one or more network access points.

[0019] Processor 140 is used to process data and / or information related to the application scenario 100 of the smart speaker. In some embodiments, processor 140 can process data, information, and / or processing results obtained from other devices or system components, and execute program instructions based on this data, information, and / or processing results to perform one or more functions described in this specification. For example, processor 140 can obtain relevant user information from user terminal 120 via network 130, obtain audio signals from smart speaker 10, and access a proportioning parameter library stored in database 150. As another example, processor can execute instructions related to generating fragrance proportioning parameters, target extraction amounts, etc., based on this data.

[0020] For more information on audio signals, fragrance formulation parameters, and target extraction volume, please refer to [link / reference]. Figure 4 And its related description. More information about the proportioning parameter library can be found at [link to relevant documentation]. Figure 6 And its related descriptions.

[0021] In some embodiments, processor 140 may include one or more sub-processing devices (e.g., a single-core processing device or a multi-core multi-chip processing device). By way of example only, processor 140 may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), a microprocessor, or any combination thereof. In some embodiments, the processor may be part of a smart speaker 110.

[0022] Database 150 can store data, instructions, and / or any other information related to the smart speaker 10. For example, database 150 can store data and / or information acquired by the smart speaker 10, user terminal 120, and processor 140 (e.g., audio signals from the smart speaker 10, fragrance blending parameters, etc.). As another example, database 150 can store current environmental data acquired by the smart speaker 10, relevant information from the user terminal 120, and a blending parameter library, etc. More information about current environmental data can be found in [link to relevant documentation]. Figure 5 And its related descriptions.

[0023] In some embodiments, database 150 may store data and / or instructions for processor 140 to perform or use in order to complete the exemplary methods described herein. For example, database 150 may store instructions for processor 140 to determine the target extraction volume of each metering pump based on fragrance blending parameters.

[0024] In some embodiments, database 150 may include one or more storage units, each of which may be a separate device or part of another device. In some embodiments, database 150 may be implemented on a cloud platform. In some embodiments, database 150 may be part of user terminal 120, processor 140, and / or smart speaker 10.

[0025] This specification provides an example of a smart speaker through some embodiments.

[0026] Figure 2 This is a structural schematic diagram of a smart speaker according to some embodiments of this specification.

[0027] In some embodiments, such as Figure 2 As shown, the smart speaker 10 includes: an audio playback module 11, a control module 12, and a fragrance blending module 13; the fragrance blending module 13 includes a base liquid storage chamber 131, a metering pump 132, a gas-liquid mixing atomization chamber 133, and an atomization driver 134; the control module 12 includes an audio analysis unit 121 and a ratio generation unit 122; the audio analysis unit 121 is configured to collect audio signals from the audio playback module and output emotion tags based on the audio signals; the ratio generation unit 122 is configured to: generate fragrance ratio parameters based on the emotion tags; determine the target extraction amount of each metering pump 132 based on the fragrance ratio parameters; control each metering pump 132 to extract the target extraction amount of fragrance base liquid from the corresponding base liquid storage chamber 131, and deliver the fragrance base liquid to the gas-liquid mixing atomization chamber 133; the atomization driver 134 is configured to mix and atomize the fragrance base liquid entering the gas-liquid mixing atomization chamber 133 for discharge.

[0028] The audio playback module 11 is used to receive electrical signals and convert them into audio signals for playback. In some embodiments, the audio playback module 11 may include at least one of a speaker, a power amplifier circuit, etc.

[0029] The control module 12 is used to process various instructions during the operation of the smart speaker 10 and control the operation of corresponding components. In some embodiments, the control module 12 integrates at least an audio analysis unit 121 and a matching generation unit 122.

[0030] In some embodiments, the control module 12 may also be located outside the smart speaker 10, such as being integrated into the processor.

[0031] The audio analysis unit 121 is used to acquire and analyze audio signals. In some embodiments, such as Figure 2 As shown, the audio analysis unit 121 may include at least a plurality of microphone arrays 1211 and an analysis unit 1212. The microphone arrays 1211 are used to acquire audio signals, and the analysis unit 1212 is used to analyze the audio signals acquired by the microphone arrays 1211. In some embodiments, the analysis unit 1212 may integrate a DSP (Digital Signal Processing) chip.

[0032] In some embodiments, the multiple microphone array 1211 can also collect the user's voice, and the analysis unit 1212 can also analyze the voice.

[0033] The proportioning generation unit 122 is used to generate fragrance proportioning parameters and control the extraction and transportation of fragrance base liquid. In some embodiments, the proportioning generation unit 122 integrates a central processing unit (MCU / SoC).

[0034] For more information on the audio analysis unit 121 and the proportion generation unit 122 mentioned above, please refer to the description below.

[0035] Fragrance blending module 13 is used to blend fragrances and output the fragrances to the external environment.

[0036] The base liquid storage chamber 131 is used to store fragrance base liquid. In some embodiments, there may be multiple base liquid storage chambers 131, and different base liquid storage chambers 131 may store different fragrance base liquids (such as different notes or different fragrance types). In some embodiments, each base liquid storage chamber 131 is connected to one end of a base liquid pipe, and the other end of the base liquid pipe is connected to the top of the gas-liquid mixing atomization chamber 133. Here, the top refers to the end of the structure that is away from the ground when the smart speaker is placed horizontally.

[0037] In some embodiments, the smart speaker 10 (such as the speaker housing 15 or the baffle 14) may be provided with multiple mounting structures (such as tanks), and multiple base liquid storage tanks 131 are detachably or fixedly installed in the corresponding mounting structures.

[0038] Metering pump 132 is used to control the amount of fragrance base liquid drawn. In some embodiments, the number of metering pumps 132 is the same as the number of base liquid storage tanks 131. In some embodiments, metering pumps 132 are disposed on the base liquid pipeline of the corresponding base liquid storage tank 131.

[0039] In some embodiments, the metering pump 132 may be a micro piezo pump or a peristaltic pump. In some embodiments, the metering pump 132 may include at least a drive unit (such as a motor or piezoelectric ceramic), a driven unit (such as a roller, pressure block or check valve), and a deformation unit (elastic hose, pump chamber or piston). When the drive unit is activated, it drives the driven unit to move, causing the deformation unit to deform and generate negative pressure, thereby drawing the fragrance base liquid from the base liquid storage tank 131.

[0040] The gas-liquid mixing atomizing chamber 133 refers to a cavity used to mix different fragrance base liquids. In some embodiments, the smart speaker 10 may further include a partition 14 disposed inside the smart speaker 10, which can be used to divide the internal space of the smart speaker 10. The gas-liquid mixing atomizing chamber 133 may be formed by the partition 14.

[0041] The atomizing driver 134 is used to atomize the mixed fragrance base liquid into micron-sized dry fog. In some embodiments, the atomizing driver 134 may be an ultrasonic atomizing plate or the like. In some embodiments, the number of atomizing drivers 134 may be one or more, and the atomizing drivers 134 are disposed at the bottom and side walls of the gas-liquid mixing atomizing chamber 133. The bottom refers to the end of the structure closest to the ground when the smart speaker is placed horizontally.

[0042] For example only, such as Figure 2 As shown, the fragrance blending module 13 includes four base liquid storage chambers 131. Each of the four base liquid storage chambers 131 stores a different fragrance base liquid (such as top notes, middle notes, base notes, or different fragrance types (such as woody, floral, fruity, marine, etc.)). The base liquid pipes of each base liquid storage chamber 131 are connected to the top of the gas-liquid mixing atomization chamber 133. Four metering pumps 132 are respectively installed on the corresponding base liquid pipes. With this design, when the user plays different music, the fragrance blended according to the music will have a different scent.

[0043] In some embodiments, the control module 12 is communicatively connected to the audio playback module 11 and the fragrance blending module 13. The audio analysis module 121 can collect the audio signal played by the audio playback module 11 and analyze the emotion represented by the audio signal; the proportioning generation unit 122 can generate fragrance proportioning parameters based on the analysis results of the audio analysis module 11, and control different metering pumps 132 of the fragrance blending module 13 to pump the corresponding fragrance base liquid, and mix, atomize and discharge multiple fragrance base liquids to blend different fragrances.

[0044] In some embodiments of this specification, the control module identifies emotions from music and accurately outputs the corresponding mixed and atomized fragrance based on those emotions, enabling the linkage of auditory and olfactory senses and creating an immersive multi-sensory experience for the user.

[0045] In some embodiments, the gas-liquid mixing atomizing chamber 133 is provided with a guide channel 1331, which is configured to physically premix a variety of fragrance base liquids before atomization.

[0046] The flow channel 1331 is used to guide and mix various fragrance base liquids. In some embodiments, there may be multiple flow channels 1331, which are stacked and inclinedly disposed between the top and bottom of the gas-liquid mixing atomizing chamber 133, and there is a gap between the end of each flow channel 1311 and at least a portion of the sidewall of the gas-liquid mixing atomizing chamber 133.

[0047] like Figure 2 As shown, when multiple fragrance base liquids flow into the top of the gas-liquid mixing atomization chamber 133 along the base liquid pipe, the multiple fragrance base liquids drip layer by layer onto each layer of guide groove 1331 under gravity and flow along the inclined guide groove 1331. During the flow, the fragrance base liquids can be physically mixed. When the mixed fragrance base liquid is atomized by the atomization driver 134 from the bottom guide groove 1331, the atomized fragrance base liquid can move along the gap between each guide groove 1331 and the side wall of the gas-liquid mixing atomization chamber 133, thereby mixing more fully and more evenly.

[0048] In some embodiments of this specification, by providing a guide groove inside the gas-liquid mixing atomization chamber, the uniformity of mixing multiple fragrance base liquids can be improved, ensuring consistent scent.

[0049] In some embodiments, such as Figure 2As shown, the smart speaker 10 also includes a speaker housing 15 and an air duct system 16. The air duct system 16 is physically isolated from the acoustic cavity 111 of the audio playback module 11. The audio playback module 11, control module 12, fragrance mixing module 13, and air duct system 16 are disposed inside the speaker housing 15. The air duct outlet 161 of the air duct system 16 is disposed on the top and / or side of the speaker housing 15. The air duct outlet 161 is used to diffuse the atomized fragrance base liquid to the external environment.

[0050] Acoustic cavity 111 refers to a sealed or tuned chamber used to optimize the playback effect of audio signals. In some embodiments, the acoustic cavity 111 of the duct system 16 and the audio playback module 11 can be separated by a partition 14, thereby achieving physical isolation.

[0051] The speaker housing 15 is used to protect the internal components of the smart speaker 10. In some embodiments, the speaker housing 15 may be made of metal, polycarbonate, or composite materials. In some embodiments, the speaker housing 15 may be rectangular, circular, or other regular or irregular shapes.

[0052] The air duct system 16 is used to deliver the atomized fragrance base liquid to the external environment. In some embodiments, the air duct system 16 may include an air duct outlet 161, an airflow device 162, an air duct 163, and an air duct inlet 164.

[0053] The air duct outlet 161 is used to discharge the atomized fragrance base liquid. In some embodiments, the air duct outlet 161 is disposed on the top and / or side shell wall of the speaker housing 15. Here, the side refers to the structural surface that surrounds and connects the top and bottom of the smart speaker 10 when it is placed.

[0054] The airflow device 162 is used to generate airflow in the air duct, and guides the atomized fragrance base liquid to diffuse through the airflow. In some embodiments, the airflow device 162 may be a fan, a blower, or the like.

[0055] The air duct inlet 164 is used to guide the atomized fragrance base liquid into the air duct 163. In some embodiments, the air duct inlet 164 may be located at the top of the gas-liquid mixing atomization chamber.

[0056] Air duct 163 refers to a channel used to guide the atomized fragrance base liquid to the external environment. In some embodiments, air duct 163 connects the gas-liquid mixing atomization chamber 133 and the external environment.

[0057] Figure 3 yes Figure 2 The diagram shows a detailed structural view of the smart speaker at point A. In some embodiments, such as... Figure 3As shown, from the air duct inlet 164 to the air duct outlet 161, the air duct 163 may include a first section 1631, a second section 1632 (i.e., a constricted section), and a third section 1633, with the airflow device 162 disposed in the third section 1633. The cross-sectional areas of the first section 1631 and the third section 1633 are larger than the cross-sectional area of ​​the second section 1632. The air duct inlet 164 is connected to the negative pressure zone of the second section 1632. According to Bernoulli's principle, when the airflow generated by the airflow device 162 passes through the second section 1632, the flow velocity increases and the pressure decreases, thus "drawing" the atomized fragrance base liquid from the gas-liquid mixing atomization chamber 133 into the air duct 163 (i.e., forming a suction effect), and then following the airflow, it is discharged from the air duct outlet 161. The cross-sectional area refers to the dimension of the air duct 163 perpendicular to the airflow direction.

[0058] In some embodiments of this specification, by physically isolating the fragrance diffusion duct from the acoustic cavity for audio playback, interference from fan vibration and airflow noise on the speaker's sound quality can be avoided, thereby ensuring high-fidelity audio output. Simultaneously, by incorporating an airflow device into the duct and creating a gentle negative pressure zone, the atomized fragrance base liquid can be evenly "drawn" into the duct, reducing the possibility of airflow turbulence and thus improving the efficiency and coverage of fragrance diffusion, enhancing the user experience.

[0059] In some embodiments, the base liquid storage chamber 131 has a pluggable snap-fit ​​structure, and a self-locking check valve is provided at the bottom of the base liquid storage chamber 131.

[0060] A pluggable snap-fit ​​structure refers to a mechanical connection structure that allows users to install or remove the base fluid storage tank 131 from the mounting structure without using tools, through operations such as pressing, rotating, or pushing and pulling. In some embodiments, the base fluid storage tank 131 and the mounting structure of the smart speaker 10 are designed with matching snaps and slots, which cooperate to enable the installation and removal of the base fluid storage tank 131.

[0061] A self-locking check valve is a one-way valve located at the bottom of the base liquid storage chamber 131 (i.e., the outlet of the fragrance base liquid). In some embodiments, the self-locking check valve can automatically close to prevent liquid leakage when the base liquid storage chamber 131 is not connected to the mounting structure and is not connected to the metering pump 132 (i.e., the base liquid storage chamber 131 is not installed to the smart speaker 10), and automatically open when the base liquid storage chamber 131 is connected to the mounting structure and connected to the metering pump 132 (i.e., the base liquid storage chamber 131 is installed to the smart speaker 10).

[0062] In some embodiments of this specification, the pluggable snap-fit ​​structure design allows users to easily replace or replenish different fragrance base liquids, improving the product's usability and maintenance convenience. The self-locking check valve at the bottom of the base liquid storage compartment prevents liquid leakage during replacement, transportation, or storage, keeping the inside of the smart speaker clean, preventing liquid corrosion of the circuitry or cross-contamination, thereby improving product reliability and lifespan.

[0063] In some embodiments, the smart speaker 10 further includes an environmental sensing module 16; the control module 12 further includes a proportioning correction unit 123, which is configured to: acquire current environmental data through the environmental sensing module; determine a volatile correction coefficient based on the current environmental data; and adjust the target extraction amount based on the volatile correction coefficient. (See also: [link to related content]) Figure 5 The description in the text.

[0064] In some embodiments, the proportioning generation unit 122 is further configured to: generate fragrance proportioning parameters based on emotion tags and a proportioning parameter library. (See also: [link to relevant content]) Figure 6 The description in the text.

[0065] In some embodiments, the control module 12 further includes a preference adaptation unit 124, which is configured to: adjust the matching parameter library according to the sentiment tags and the feedback information in response to receiving user feedback information; the matching parameter library includes sentiment tags and their corresponding preset matching parameters. The preference adaptation unit 124 is used to adaptively adjust the preset matching parameters according to user preferences. For more details on this part, please refer to [link to relevant documentation]. Figure 6 The description in the text.

[0066] In some embodiments, when the audio playback module 11 plays an audio signal, the audio analysis unit 121 can collect and analyze the audio signal to obtain the emotional tag corresponding to the audio signal; the proportioning generation unit 122 can generate fragrance proportioning parameters based on the emotional tag and control multiple metering pumps 132 to draw the corresponding fragrance base liquid from the corresponding base liquid storage tank 131 into the gas-liquid mixing atomization chamber 133; the airflow device 162 is activated to generate airflow and apply negative pressure suction to the gas-liquid mixing atomization chamber 133, and multiple fragrance base liquids flow and mix under the combined action of gravity and negative pressure suction; the atomization driver 134 is activated to atomize the mixed fragrance base liquid; the atomized fragrance base liquid enters the air duct 163 under negative pressure suction and is discharged to the external environment from the air duct outlet 161. For more details on this part, see Figures 4-6 The relevant description in the document.

[0067] This specification also provides a method for controlling a smart speaker.

[0068] Figure 4 This is an exemplary schematic diagram of a control method for a smart speaker according to some embodiments of this specification. In some embodiments, process 400 may be executed by processor 140 or control module 12.

[0069] Step 410: Collect the audio signal from the audio playback module and output the emotion tag based on the audio signal.

[0070] An audio signal refers to the sound signal played by the audio playback unit. For example, an audio signal can be music played by the audio playback unit. In some embodiments, the control module can acquire the audio signal from the audio playback module.

[0071] Emotional tags refer to semantic classification identifiers of the emotional types contained in an audio signal. In some embodiments, the emotional tag of an audio signal can be a set of multiple sub-emotional tags corresponding to multiple acoustic features of the audio signal.

[0072] In some embodiments, the control module can perform spectral analysis (such as Fourier transform) on the audio signal to determine its signal spectrum and thus its acoustic characteristics; and determine the emotion tag corresponding to the audio signal based on a pre-established "acoustic feature-emotion" mapping model (such as Russell's emotion loop model).

[0073] Acoustic features are used to characterize the acoustic properties of an audio signal. In some embodiments, acoustic features may include at least the rhythm (e.g., fast or slow), loudness (e.g., high or low loudness), and tonality (e.g., major or gentle harmony) of each segment of the audio signal.

[0074] In some embodiments, the control module can detect abrupt changes in acoustic energy in the signal spectrum (such as points where the acoustic energy change exceeds a change threshold), and convert the frequency of these abrupt changes into beats per minute (BPM) using a first preset algorithm (such as autocorrelation, beat tracking, etc.). If the BPM of a segment is greater than the first beat threshold, the audio signal of that segment is marked as "fast-paced"; if the BPM of a segment is less than the second beat threshold, the audio signal of that segment is marked as "slow-paced". The change threshold, the first beat threshold, and the second beat threshold are determined by a technician based on experience, with the first beat threshold being greater than the second beat threshold.

[0075] In some embodiments, the control module may mark segments of the signal spectrum with acoustic energy greater than a first acoustic energy threshold as high loudness and segments with acoustic energy less than a second acoustic energy threshold as low loudness. The first and second acoustic energy thresholds are determined by a technician based on experience, and the first acoustic energy threshold is greater than the second acoustic energy threshold.

[0076] In some embodiments, the control module can determine a chromatogram based on the signal spectrum using a second preset algorithm (such as constant Q-transform); and match each segment of the chromatogram with a preset tone profile to determine the tone of each segment. The preset tone profile may include multiple tones and can be determined by user input. For example, the preset tone profile may be a Krumhansl-Schmuckler tone profile.

[0077] In other embodiments, the control module may also determine the acoustic characteristics of the audio signal in other ways, such as through a trained machine learning model.

[0078] In some embodiments, for a given audio signal, the control module can map the fast tempo, high loudness, and major key acoustic features extracted from the first sub-audio signal to the "high arousal-high pleasure" quadrant in Russell's emotion loop model and output a sub-emotion label of "happy / exhilarating". Similarly, based on the slow tempo, low loudness, and smooth harmony acoustic features extracted from the second sub-audio signal, it can map these features to the "low arousal-high pleasure" quadrant in Russell's emotion loop model and output a sub-emotion label of "relaxed / soothing". Therefore, the emotion label for the audio signal is "exhilarating / happy, relaxed / soothing". The sub-audio signals can be divided in various ways, such as based on preset duration or differences in acoustic features. For example, based on a preset duration, the last segment shorter than the preset duration is retained as an independent sub-audio signal. The correspondence between acoustic features and sub-emotion labels can be preset based on experience.

[0079] In some embodiments, the control module may also determine the emotion tag based on the audio signal in other ways; for example, by querying historical records, when the audio signal or an audio signal with a similarity greater than a preset threshold is played in the past, the emotion tag set by the user is set as the emotion tag for the current playback. The similarity of the audio signals can be determined by matching based on the similarity of acoustic features, etc.

[0080] Step 420: Generate fragrance blending parameters based on emotion tags.

[0081] Fragrance blending parameters refer to the set of proportions (e.g., volume) of fragrance base liquid extracted from different base liquid storage chambers relative to the total extracted amount, in order to adapt to the atmosphere corresponding to a certain sub-emotional label in the emotional label. In some embodiments, fragrance blending parameters may include multiple blending sub-parameters, where each blending sub-parameter represents the proportion (e.g., volume) of fragrance base liquid extracted from a corresponding base liquid storage chamber relative to the total extracted amount.

[0082] As an example only, when three base liquid storage chambers are set up (i.e., base liquid storage chamber a, base liquid storage chamber b, and base liquid storage chamber c), the fragrance formulation parameter R can be { }, proportion sub-parameters { } indicates that the proportion of fragrance base liquid extracted from base liquid storage chamber a is 25% of the total extracted amount, and the proportioning sub-parameter { The proportion of fragrance base liquid extracted from base liquid storage chamber b is 40% of the total extracted amount, and the proportioning sub-parameters { } indicates that the amount of fragrance base liquid extracted from the base liquid storage chamber c accounts for 35% of the total extraction amount.

[0083] In some embodiments, the control module can also generate fragrance blending parameters based on emotion tags and a blending parameter library. (See also: [link to related content]). Figure 6 The relevant description in the document.

[0084] Step 430: Determine the target extraction volume of each metering pump based on the fragrance mixing parameters.

[0085] The target extraction volume refers to the volume of fragrance base liquid that the metering pump needs to extract from each base liquid storage compartment in order to perform one fragrance base liquid release corresponding to a sub-emotion tag.

[0086] In some embodiments, the control module can calculate the target extraction amount based on the fragrance ratio parameters, the total volume of the fragrance base liquid released in a single release, and the concentration of the fragrance base liquid. As an example only, the ratio generation module can calculate the target extraction amount using the following formula (1).

[0087] (1), in, This represents the target extraction volume of the i-th metering pump; This represents the proportioning sub-parameter corresponding to the fragrance base liquid extracted by the i-th metering pump in the fragrance proportioning parameters; This indicates the total volume of fragrance base liquid released in a single batch (i.e., the total amount extracted).

[0088] In some embodiments, the fragrance blending parameters may also include the total extraction volume, which may be the same or different depending on the different emotional tags. The total extraction volume may be a system default value, an empirical value, a manually preset value, or any combination thereof. As an example only, when the sub-audio signal is divided based on a preset duration of 1 minute, the total extraction volume for each instance may be the same, such as 0.1 ml for each instance.

[0089] In some embodiments, the concentration of the fragrance base liquid stored in the base liquid storage chamber can be the same (i.e., all are standard concentrations).

[0090] In some embodiments, the concentrations of fragrance base liquids stored in the base liquid storage tanks may vary. In this case, the control module can adjust the corresponding target extraction volume based on the concentration of the fragrance base liquid. For example, the concentration of the fragrance base liquid is negatively correlated with its corresponding target extraction volume. The concentration of the fragrance base liquid can be determined by user input.

[0091] Step 440: Control each metering pump to extract the target amount of fragrance base liquid from the corresponding base liquid storage chamber, and deliver the fragrance base liquid to the gas-liquid mixing atomization chamber.

[0092] In some embodiments, the control module can control the metering pump to extract the target amount of fragrance base liquid from the corresponding base liquid storage tank, and multiple fragrance base liquids are transported to the top of the gas-liquid mixing atomization chamber through the corresponding base liquid pipeline.

[0093] Step 450: Control the atomizing driver to mix and atomize the fragrance base liquid entering the gas-liquid mixing atomizing chamber for discharge.

[0094] In some embodiments, multiple fragrance base liquids drip layer by layer into each layer of guide channels under gravity and flow along the inclined guide channels. During the flow, the fragrance base liquids can be mixed. When the mixed fragrance base liquid is atomized by the atomizing driver from the bottom guide channel, the atomized fragrance base liquid can be drawn into the air duct along the gap between each guide channel and the side wall of the gas-liquid mixing atomization chamber under the action of the airflow formed by the airflow device and the negative pressure formed by the air duct, and discharged to the external environment from the air duct outlet.

[0095] In some embodiments of this specification, the audio analysis module identifies emotions from music, and the fragrance blending module accurately outputs the corresponding mixed and atomized fragrance based on the emotions, which can realize the linkage of hearing and smell, creating an immersive multi-sensory experience for users.

[0096] It should be noted that the above description of process 400 is for illustrative purposes only and does not limit the scope of this specification. Those skilled in the art can make various modifications and changes to process 400 under the guidance of this specification. However, these modifications and changes remain within the scope of this specification.

[0097] In some embodiments, when the emotion label changes, such as when the coordinate values ​​of acoustic features in Russell's emotion loop model change beyond a threshold (i.e., a significant change in music genre), the control module can empty the fragrance base liquid in the base liquid storage chamber within a preset time before executing fragrance extraction control based on the new emotion label (e.g., by spraying an odorless solvent or using airflow to remove residual fragrance) to prevent odor cross-contamination. The configuration and structure of the odorless solvent storage chamber are similar to those of the base liquid storage chamber. The change threshold and preset time are set based on actual needs.

[0098] Figure 5 This is an exemplary flowchart illustrating the adjustment of the target extraction amount according to some embodiments of this specification. Figure 5 As shown, process 500 includes the following steps. In some embodiments, process 500 may be executed by processor 140 or control module 12.

[0099] In some embodiments, the smart speaker 10 further includes an environmental sensing module; the control module further includes a ratio correction unit 122. More information about the control module can be found in [link to relevant documentation]. Figure 2 And its related descriptions.

[0100] An environmental sensing module is used to sense data related to the environment. In some embodiments, the environmental sensing module can be disposed on the surface of the smart speaker's casing and may include a humidity sensor, a temperature sensor, etc.

[0101] The proportioning correction unit is used to correct the target extraction amount. In some embodiments, the proportioning correction unit may be located on the main control circuit board of the smart speaker 10. More information about the main control circuit board can be found in [link to relevant documentation]. Figure 2 And its related description. For more information on the target extraction volume, please refer to... Figure 4 And its related descriptions.

[0102] Step 510: Obtain current environmental data through the environmental perception module.

[0103] Current environmental data refers to data related to the current environment in which the smart speaker is located. In some embodiments, current environmental data may include the current temperature, humidity, etc.

[0104] In some embodiments, the processor can obtain the current temperature and humidity of the environment through the temperature sensor and humidity sensor in the environmental sensing module, and use the temperature and humidity as the current environmental data.

[0105] Step 520: Determine the volatility correction factor based on the current environmental data.

[0106] The volatility correction factor refers to the proportion by which the target extraction amount needs to be adjusted to achieve the same expected olfactory perception intensity as under standard experimental conditions, after taking into account the impact of current environmental data.

[0107] The standard experimental environment refers to the baseline environment that does not consider the impact of different environmental data on fragrance volatilization and diffusion.

[0108] In some embodiments, the temperature range of the standard experimental environment can be between 20°C and 25°C, and the humidity range can be between 40% and 60%. The temperature and humidity ranges of the standard experimental environment can be determined according to specific needs and are not limited here.

[0109] Expected olfactory perception intensity refers to the intensity of the aroma that a user expects to perceive under standard experimental conditions.

[0110] In some embodiments, the expected olfactory perception intensity is set as a uniform benchmark value so that the intensity of the aroma perceived by the user in different environments can reach or approach the uniform expected olfactory perception intensity.

[0111] In some embodiments, the processor can determine the expected olfactory perception intensity in a variety of ways. For example, in a standard experimental environment, a professional sensory evaluation team assesses the olfactory sensation in the standard experimental environment and, based on preset comfort standards, jointly determines a suitable aroma intensity as a unified benchmark. This aroma intensity is then determined as the expected olfactory perception intensity in the standard experimental environment.

[0112] In some embodiments, the processor calculates a volatility correction coefficient based on the expected olfactory perception intensity and current environmental data, and applies a unified logical correction to the target extraction amount corresponding to any fragrance blending parameter. More information about the volatility correction coefficient can be found below and in its related description.

[0113] In some embodiments, technicians can determine the volatility correction factor in various ways based on current environmental data.

[0114] In some embodiments, the processor can empirically set a conventional volatile correction factor (e.g., 1). If the processor detects that the current ambient temperature is higher than the maximum value of the temperature range in the standard experimental environment and the humidity is lower than the minimum value of the humidity range in the standard experimental environment, it indicates that the molecular thermal motion in the current environment is fast. At this time, the aroma perception is usually stronger. The processor automatically reduces the preset value, such as 20%, based on the conventional volatile correction factor, and the corrected factor is the volatile correction factor. At this time, the volatile correction factor is less than 1 (e.g., 0.8). Conversely, the processor can automatically increase the preset value based on the conventional volatile correction factor, and the corrected factor is the volatile correction factor. At this time, the volatile correction factor is greater than 1 (e.g., 1.2).

[0115] In some embodiments, there are multiple ways to set the volatile correction coefficient. For example, it can be based on a coefficient lookup table, which contains various environmental data and their corresponding volatile correction coefficients preset based on historical experience. Alternatively, based on the relationship that higher temperature results in a smaller volatile correction coefficient, and higher humidity results in a larger volatile correction coefficient, a volatile correction coefficient can be determined by weighting temperature and humidity, with temperature having a greater weight than humidity.

[0116] In some embodiments, the processor can also determine a volatility correction factor based on the fragrance blending parameters and current environmental data.

[0117] For more information on fragrance formulation parameters, please refer to [link / reference]. Figure 4 And its related descriptions.

[0118] For example, the processor can determine the volatility correction coefficient based on the fragrance formulation parameters and current environmental data through a correction model.

[0119] In some embodiments, the input to the correction model may include fragrance blending parameters and current environmental data, and the output may include a volatility correction coefficient.

[0120] In some embodiments, the modified model can be trained based on a large number of training samples with training labels. The processor can input multiple training samples with training labels into the initial modified model, construct a loss function using the training labels and the results of the initial modified model, and iteratively update the parameters of the initial modified model based on the loss function using methods such as gradient descent. When the loss function meets preset training conditions, the trained modified model is obtained. These preset training conditions may include loss function convergence, the number of iterations reaching a threshold, etc.

[0121] The training samples include the corresponding fragrance blending parameters and historical environmental data for historical fragrance spraying events. The training labels are the actual volatilization correction coefficients for the historical fragrance spraying events. Both the training samples and training labels can be obtained based on historical data. The historical fragrance spraying events can be selected from those with positive user feedback or those where no user feedback on parameter adjustments was received.

[0122] As an example, the processor can filter historical fragrance spraying records from historical data that have not been adjusted by the user within a short period (e.g., 5 minutes) as sample data; the fragrance blending parameters and current environmental data in the sample data are used as training samples; based on the actual target extraction amount corresponding to the training sample, the processor calculates the corresponding actual evaporation correction coefficient by using the relationship between the actual evaporation correction coefficient and the ratio of the actual target extraction amount to the target extraction amount before correction, and uses this actual evaporation correction coefficient as the training label corresponding to the training sample. Here, "no user adjustment" means that the user has not manually adjusted the target extraction amount, fragrance blending parameters, etc.

[0123] The actual target extraction amount corresponding to the training sample refers to the amount of fragrance that can be extracted to meet the expected olfactory perception intensity under the sample fragrance ratio parameters and historical environmental data corresponding to historical fragrance events. The processor can determine the actual target extraction amount corresponding to each fragrance base liquid based on the actual extraction situation of each fragrance base liquid in historical fragrance events. The processor can pre-set the extraction amount of each metering pump that can achieve the expected olfactory perception intensity under standard experimental conditions for each fragrance ratio parameter based on the volatility characteristics of the fragrance in the fragrance ratio parameter, and use the extraction amount of each metering pump as its corresponding target extraction amount before correction.

[0124] Some embodiments in this specification combine fragrance ratio parameters (such as the volatility, diffusion, and sensory intensity differences of different fragrance types) to achieve personalized compensation adjustments for each fragrance ratio parameter. This specification abandons the approach of using a uniform correction logic for all fragrance ratio parameters, avoiding the problem of some fragrances being too strong or too weak due to uniform correction. This allows fragrance adjustment to adapt to the characteristic needs of different fragrance ratio parameters, further enhancing the intelligence and accuracy of the fragrance experience, and ensuring that the fragrances of different fragrance ratio parameters can present the expected sensory effects in the environment.

[0125] Step 530: Adjust the target extraction amount according to the volatility correction coefficient.

[0126] In some embodiments, the processor can adjust the target extraction amount based on the volatility correction factor. For example, the processor can adjust the target extraction amount using the following formula (2) to obtain the corrected target extraction amount: (2) in, This represents the target extraction volume of the i-th metering pump before correction. See details below. Figure 4 The corresponding content; This represents the volatile correction factor corresponding to the i-th metering pump; This represents the target extraction volume of the i-th metering pump after correction by the volatility correction factor.

[0127] Some embodiments in this specification address the interference of environmental factors on the aroma diffusion effect by adding an environmental sensing module and a proportioning correction unit. Through the linkage adjustment of current environmental data and the volatility correction coefficient, the processor can automatically adapt the target extraction amount of the fragrance base liquid according to the actual environment, ensuring that the olfactory perception intensity of the aroma remains stable and consistent with expectations under different temperature and humidity conditions. This effectively improves the environmental adaptability of the smart speaker's aroma function, upgrading the aroma experience from simple content matching to a more precise presentation tailored to actual scenarios.

[0128] Figure 6 This is a schematic diagram illustrating the fragrance formulation parameters according to some embodiments of this specification.

[0129] In some embodiments, such as Figure 6 As shown, the ratio generation unit is further configured to generate fragrance ratio parameters 630 based on the emotion tag 610 and the ratio parameter library 620.

[0130] A blending parameter library is a database containing the correspondence between multiple reference emotional tags and preset blending parameters. Preset blending parameters refer to the fragrance blending parameters set in the blending parameter library that are available for use.

[0131] In some embodiments, the blending generation unit can query a blending parameter library based on the emotion tag to determine a reference emotion tag that meets preset requirements, and use the preset blending parameters corresponding to the reference emotion tag as the fragrance blending parameters for that emotion tag. The preset requirements are determined based on experience. For example, the preset requirements could be: in Russell's emotion loop model, the distance between the coordinates of the emotion tag and the coordinates of the reference emotion tag is less than a distance threshold, etc. The distance threshold is determined based on experience.

[0132] In some embodiments, the proportioning parameter library can be constructed based on experiments. For example, technicians construct a reference emotional tag and create multiple fragrances, each corresponding to a different reference fragrance proportioning parameter. A predetermined number of users are then allowed to smell the fragrances to select the most satisfactory one. The reference fragrance proportioning parameter corresponding to the fragrance selected most frequently by the users is used as the preset proportioning parameter for that emotional tag. Through this method, multiple emotional tags and their corresponding preset proportioning parameters can be determined, thus forming a proportioning parameter library. The predetermined number is determined based on actual needs.

[0133] In the embodiments of this specification, by setting up a ratio parameter library, the mapping between emotional tags and fragrance ratio parameters can be realized, enabling the system to efficiently query verified high-quality formulas, greatly reducing real-time computational complexity and improving system response speed and stability.

[0134] In some embodiments, the control module further includes a preference adaptation unit, such as Figure 6 As shown, the preference adaptive unit is configured to: in response to receiving feedback information 640 from the user, adjust the matching parameter library 620 according to the sentiment tag 610 and the feedback information 640; the matching parameter library 620 includes sentiment tags and their corresponding preset matching parameters.

[0135] Feedback information refers to the user's evaluation of the currently used fragrance blend parameters. For example, feedback information can include negative feedback (such as "the woody scent is too strong" or "the floral scent is too weak") and positive feedback (such as "I like this blend" or "Save this blend"). In some embodiments, feedback information can be obtained through the user's voice captured by a microphone array and recognized by the analysis unit using a large language model. The analysis unit can then send the feedback information to the preference adaptation unit.

[0136] In some embodiments, the preference adaptive unit can obtain multiple preset proportioning parameters corresponding to emotion tags in the proportioning parameter library based on emotion tags and feedback information. Within these preset proportioning parameters, it can decrease the value of the proportioning sub-parameter corresponding to fragrance base liquids that the user dislikes, and increase the value of the proportioning sub-parameter corresponding to the remaining fragrance base liquids, generating multiple adjusted fragrance proportioning parameters and updating the proportioning parameter library. The magnitude of the increase and decrease in the value of the proportioning sub-parameters is based on empirical settings.

[0137] As an example, if a user inputs "the woody scent is too bitter" when playing an audio with a melancholic emotional tag, the preference adaptive unit can query the blending parameter library based on the emotional tag and the feedback information, and find two preset blending parameters with the corresponding emotional tag "melancholy". The preference adaptive unit will reduce the value of the blending sub-parameter corresponding to the fragrance base liquid (and woody scent) that the user dislikes, and increase the value of the blending sub-parameter corresponding to other fragrance base liquids (such as floral, citrus, etc.), generate two corresponding adjusted fragrance blending parameters, and update the blending parameter library.

[0138] In some embodiments of this specification, collecting user feedback information to adjust various preset proportion parameters with the same emotional label can make the fragrance gradually match the user's personal preferences, enabling the product to have the ability to continuously improve and personalize, thereby enhancing user stickiness.

[0139] In some embodiments, the multiple preset ratio parameters in the ratio parameter library are further configured with preference weights, and the preference adaptive unit is further configured to adjust the preference weights according to multiple historical feedback information of the user and the corresponding multiple historical ratio parameters every preset period.

[0140] In some embodiments, since the preference adaptation unit can adjust the preset ratio parameters corresponding to the emotion tag based on user feedback when using the smart speaker, the preference adaptation unit can retain multiple feasible (e.g., those used by the user) preset ratio parameters (i.e., an emotion tag can have multiple preset ratio parameters) and assign initial preference weights to the preset ratio parameters. The initial preference weights of multiple preset ratio parameters can be equal or preset. The preference weights are used to measure the priority of a user liking or accepting a certain preset ratio parameter.

[0141] Historical feedback information refers to the feedback information entered by the user before the current moment. Historical blending parameters refer to the fragrance blending parameters used by the user before the current moment. Preset period refers to the period during which preference weights are adjusted.

[0142] In some embodiments, in response to reaching a preset period (e.g., the time interval since the last update of the proportioning parameter library reaches a preset period), the preference adaptive unit can select historical feedback information and historical proportioning parameters from the previous N preset periods at the current time. For each preset proportioning parameter corresponding to a historical proportioning parameter, the preference adaptive unit can adjust its preference weight in various ways. The preset period and N can be system default values, empirical values, manually preset values, or any combination thereof.

[0143] For example, the preference adaptive unit can, based on historical feedback information and historical matching parameters, reduce the preference weight of the preset matching parameter corresponding to a number of negative feedback messages exceeding a first quantity threshold, and increase the preference weight of the preset matching parameter when the number of uses exceeds a usage threshold and the number of negative feedback messages is less than the first quantity threshold. The usage threshold and the first quantity threshold can be system default values, empirical values, manually preset values, or any combination thereof.

[0144] For example, the preference adaptive unit can count the number of positive feedback messages and the number of negative feedback messages in the historical feedback information to determine the preference sub-weight of each ratio sub-parameter; based on the preference sub-weights of multiple ratio sub-parameters, it can determine the preference weight corresponding to the preset ratio parameter.

[0145] Preference sub-weights are used to measure the priority of a certain ratio sub-parameter in the preset ratio parameters in terms of user preference or acceptance.

[0146] In some embodiments, the preference adaptation unit can determine the preference sub-weight of the matching sub-parameter based on its utility value. The utility value reflects the user's degree of liking for the matching sub-parameter.

[0147] As an example only, the utility value of the ratio sub-parameter can be determined by the following formula (3).

[0148] (3), in, This represents the utility value of the i-th ratio sub-parameter; This represents the total number of positive feedback messages in the historical feedback information of the previous N preset periods at the current moment; This represents the total number of negative feedback messages in the historical feedback information of the previous N preset periods at the current moment; This indicates the number of times the preset ratio parameter containing the i-th ratio sub-parameter has been fed back in the historical feedback information of the previous N preset periods at the current moment; This represents the number of times each preset ratio parameter was fed back in the historical feedback information of the previous N preset periods at the current moment. and It can be determined based on historical feedback data and historical proportioning parameters.

[0149] In some embodiments, the preference sub-weights of the matching sub-parameters are positively correlated with their utility values, and the preference adaptation unit can determine the corresponding preference sub-weights based on the utility values ​​of each matching sub-parameter.

[0150] In some embodiments, the preference adaptive unit can perform a weighted summation of the preference sub-weights of all the sub-parameters of the preset ratio parameter to obtain the preference weight of the preset ratio parameter. The weights are set based on experience.

[0151] In some embodiments, for a certain emotion tag, the ratio generation unit can select the preset ratio parameter corresponding to the preference weight with the largest value under that emotion tag as the fragrance ratio parameter, and then mix and atomize the fragrance base liquid.

[0152] In some embodiments of this specification, by periodically analyzing user usage information and feedback information, not only can the smart speaker automatically and progressively optimize its formula selection strategy and increase its intelligence level, but it can also uncover deeper preference patterns from the long-term behavioral patterns of user groups, improve the overall recommendation quality of the system, and thus enhance the user experience.

[0153] The basic concepts have been described above. Obviously, for those skilled in the art, the detailed disclosure above is merely illustrative and does not constitute a limitation of this specification. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements, and corrections to this specification. Such modifications, improvements, and corrections are suggested in this specification and therefore remain within the spirit and scope of the exemplary embodiments described herein.

[0154] Finally, it should be understood that the embodiments in this specification are merely illustrative of the principles of the embodiments described herein. Other variations may also fall within the scope of this specification. Therefore, alternative configurations of the embodiments in this specification are intended to be illustrative rather than limiting, and should be considered consistent with the teachings of this specification. Accordingly, the embodiments in this specification are not limited to those explicitly described and illustrated herein.

Claims

1. A smart speaker, characterized in that, include: Audio playback module, control module, fragrance blending module; The fragrance blending module includes a base liquid storage tank, a metering pump, a gas-liquid mixing atomizing chamber, and an atomizing driver. The control module includes an audio analysis unit and a proportion generation unit; The audio analysis unit is configured to acquire the audio signal from the audio playback module and output an emotion tag based on the audio signal. The proportioning generation unit is configured as follows: Based on the emotional tags, generate fragrance blending parameters; Based on the fragrance formulation parameters, the target extraction volume of each metering pump is determined; Each metering pump is controlled to draw the target amount of fragrance base liquid from the corresponding base liquid storage chamber and deliver the fragrance base liquid to the gas-liquid mixing atomization chamber; The atomizing driver is configured to mix and atomize the fragrance base liquid entering the gas-liquid mixing atomizing chamber for discharge.

2. The smart speaker according to claim 1, characterized in that, It also includes an environmental perception module; The control module further includes a proportioning correction unit, which is configured to: The current environmental data is obtained through the environmental perception module. Based on the current environmental data, determine the volatility correction factor; The target extraction amount is adjusted based on the volatility correction factor.

3. The smart speaker according to claim 1, characterized in that, The proportioning generation unit is further configured as follows: Based on the emotional tags, fragrance blending parameters are generated using a blending parameter library.

4. The smart speaker according to claim 3, characterized in that, The control module further includes a preference adaptation unit, which is configured to: In response to receiving user feedback, the system adjusts the matching parameter library based on the emotion tags and the feedback information; the matching parameter library includes the emotion tags and their corresponding preset matching parameters.

5. The smart speaker according to claim 1, characterized in that, The gas-liquid mixing atomization chamber is provided with a flow guide groove, which is configured to physically premix the various fragrance base liquids before atomization.

6. The smart speaker according to claim 1, characterized in that, It also includes a speaker housing and an air duct system, the air duct system being physically isolated from the acoustic cavity of the audio playback module; the audio playback module, the control module, the fragrance mixing module, and the air duct system are housed within the speaker housing. The air duct outlet of the air duct system is located on the top and / or side of the speaker housing, and the air duct outlet is used to diffuse the atomized fragrance base liquid into the external environment.

7. The smart speaker according to claim 1, characterized in that, The base liquid storage chamber has a pluggable snap-fit ​​structure, and a self-locking check valve is provided at the bottom of the base liquid storage chamber.

8. A control method for a smart speaker, characterized in that, include: The system acquires audio signals from the audio playback module and outputs emotion tags based on the audio signals. Based on the emotional tags, generate fragrance blending parameters; Based on the fragrance formulation parameters, determine the target extraction volume of each metering pump; Each metering pump is controlled to draw the target amount of fragrance base liquid from the corresponding base liquid storage chamber and deliver the fragrance base liquid to the gas-liquid mixing atomization chamber; The atomizing driver is controlled to mix and atomize the fragrance base liquid entering the gas-liquid mixing atomizing chamber for discharge.

9. A control device for a smart speaker, the device comprising at least one processor and at least one memory; The at least one memory is used to store computer instructions; The at least one processor is configured to execute at least a portion of the computer instructions to implement the control method as described in claim 8.

10. A computer-readable storage medium storing computer instructions that, when executed by a processor, implement the control method of claim 8.

Images