Automatic sound conversion device

Abstract

The invention discloses an automatic sound conversion device, and relates to the technical field of sound recognition. The device comprises an adapter, a server and a sprayer which are connected in sequence, wherein the adapter is used for collecting a sound signal and sending the sound signal to the server; the server is used for identifying a scene according to the sound signal, mapping the scene into a corresponding smell code, and sending the smell code to the sprayer; the sprayers are used for spraying corresponding smells according to the smell codes. According to the automatic sound conversion device disclosed by the embodiment of the invention, different essential oils can be selected according to different scenes and different requirements, so that different smells are generated. The device is simple in structure and convenient to carry, can be widely applied to home entertainment, network caffees, game halls, cinemas, school teaching and the like, and brings better user experience to consumers.

Description

Technical Field

[0001] The present invention relates to the technical field of sound recognition, and in particular to an automatic sound conversion device. Background Art

[0002] Now many places have launched 4D or 5D cinemas, which add physical senses to the traditional cinemas with only vision and hearing, such as movable seats, spray or water. At the same time, there are many VR / AR devices that can make users feel as if they are in the game, which enhances the gaming experience. Although these technologies can enhance the user's experience of scenes such as videos / games, most of the equipment used in these technologies is relatively expensive and costly, and is limited to specific places, so ordinary consumers cannot experience it at home. Summary of the invention

[0003] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention proposes a sound automatic conversion device, which can improve user experience in a low-cost manner.

[0004] According to an embodiment of the present invention, the automatic sound conversion device includes: an adapter, a server and a sprayer connected in sequence; wherein the adapter is used to collect sound signals and send the sound signals to the server; the server is used to identify scenes according to the sound signals, and map the scenes to corresponding odor codes, and send the odor codes to the sprayer; the sprayer is used to spray corresponding odors according to the odor codes.

[0005] According to some embodiments of the present invention, the adapter includes multiple sound transmission paths, a sound switching module and a sound output module, the multiple sound transmission paths are used to receive sound signals from different interfaces, the sound switching module is used to select the sound signals of the multiple sound transmission paths, and send the selected sound signals to the server through the sound output module.

[0006] According to some embodiments of the present invention, one of the sound transmission paths comprises:

[0007] An optical fiber input interface, used for receiving a first sound signal;

[0008] A signal receiver, configured to amplify the first sound signal;

[0009] The audio decoding chip is used to decode the amplified first sound signal to obtain a first I2S signal, and send the first I2S signal to the sound switching module.

[0010] According to some embodiments of the present invention, one of the sound transmission paths comprises:

[0011] An HDMI input interface, used for receiving a first audio and video signal;

[0012] The HDMI processing chip is used to extract the audio information of the first audio and video signal and output a second I2S signal to the sound switching module.

[0013] According to some embodiments of the present invention, the HDMI processing chip is further connected to the optical fiber output interface via a first switching switch, and the optical fiber input interface is further connected to the optical fiber output interface via the first switching switch.

[0014] According to some embodiments of the present invention, one of the sound transmission paths comprises:

[0015] An HDMI output interface is used to receive a second audio and video signal, and send the second audio and video signal to the HDMI processing chip, so that the HDMI processing chip outputs a third I2S signal to the sound switching module. The HDMI output interface is also connected to the first switching switch through the HDMI processing chip, and then connected to the optical fiber output interface through the first switching switch.

[0016] According to some embodiments of the present invention, one of the sound transmission paths comprises:

[0017] A headphone input interface, used for receiving a second sound signal;

[0018] The analog-to-digital conversion chip is used to convert the second sound signal into a pcm format and output a fourth I2S signal to the sound switching module.

[0019] According to some embodiments of the present invention, the sound switching module includes:

[0020] A second switching switch is used to switch the sound signals output by the plurality of sound transmission paths and select a corresponding sound signal;

[0021] A sound splitting unit, used for dividing the sound signal into two paths, and sending one of the sound signals to the server through the sound output module;

[0022] A digital-to-analog conversion chip, used for performing digital-to-analog conversion on another channel of the sound signal to obtain a third sound signal;

[0023] The third switch is used to select between the second sound signal and the third sound signal, and output the selected signal to the headphone output interface.

[0024] According to some embodiments of the present invention, the server comprises:

[0025] Wowza streaming media server, the adapter sends audio stream data to the Wowza streaming media server according to the sound signal;

[0026] Kafka, the Wowza streaming server sends the audio stream data to the Kafka through a message queue;

[0027] Lambda, obtains the audio stream data by subscribing to the Kafka message, and adjusts the window size of the audio stream data to a preset window;

[0028] SageMaker receives the audio stream data of the preset window, performs analysis and event inference, obtains a recognition result of the audio stream data, and sends the recognition result to Lambda, and Lambda sends the recognition result to the message queue of Kafka;

[0029] The business server is used to output the corresponding odor code to the sprayer according to the recognition result of the Kafka message queue.

[0030] According to some embodiments of the present invention, the business server also communicates with AWS IoT Core, and manages and controls the adapter and the sprayer through the AWS IoT Core.

[0031] The automatic sound conversion device according to the embodiment of the present invention has at least the following beneficial effects: different essential oils can be selected according to different scenes and different needs, thereby producing different smells. The device has a simple structure and is easy to carry, and can be widely used in home entertainment, Internet cafes, game rooms, cinemas, school teaching, etc., giving consumers a better user experience. The device can identify different scenes according to different sounds, thereby releasing different smells, allowing consumers to have a better experience, especially for game enthusiasts. During the game, they can also experience various smells such as car exhaust, gun smoke, blood, etc., making the game experience better. Since the sound is converted into smell, the olfactory experience is increased, which adds a new experience to the audience watching movies, TV and playing games. In addition to vision and hearing, the sense of smell is also increased.

[0032] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The above and / or additional aspects and advantages of the present invention will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which:

[0034] Figure 1A schematic diagram of the structure of an automatic sound conversion device according to an embodiment of the present invention;

[0035] Figure 2 is a schematic diagram of one of the sound transmission paths of an embodiment of the present invention;

[0036] Figure 3 is a schematic diagram of one of the sound transmission paths of an embodiment of the present invention;

[0037] Figure 4 is a schematic diagram of one of the sound transmission paths of an embodiment of the present invention;

[0038] Figure 5 is a schematic diagram of one of the sound transmission paths of an embodiment of the present invention;

[0039] Figure 6 Schematic diagram of a sprayer according to an embodiment of the present invention. DETAILED DESCRIPTION

[0040] Embodiments of the present invention are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application, and are not to be construed as limitations on the present application. For the step numbers in the following embodiments, they are only provided for the convenience of explanation, and the order between the steps is not limited in any way, and the execution order of each step in the embodiment can be adaptively adjusted according to the understanding of those skilled in the art.

[0041] In the description of the present invention, it should be understood that descriptions involving orientations, such as up, down, front, back, left, right, etc., and orientations or positional relationships indicated are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present invention.

[0042] The terms "first", "second", "third" and "fourth" etc. in the specification and claims of the present invention and the drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or devices.

[0043] Reference to "embodiments" in the present invention means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0044] Now many places have launched 4D or 5D cinemas, which add physical senses to the traditional cinemas with only vision and hearing, such as movable seats, spray or water. At the same time, there are many VR / AR devices that can make users feel as if they are in the game, which enhances the gaming experience. Although these technologies can enhance the user's experience of scenes such as videos / games, most of the equipment used in these technologies is relatively expensive and costly, and is limited to specific places, so ordinary consumers cannot experience it at home.

[0045] To this end, an embodiment of the present invention provides an automatic sound conversion device, which can select different essential oils according to different scenes and different needs, thereby producing different smells. The device has a simple structure and is easy to carry. It can be widely used in home entertainment, Internet cafes, game halls, cinemas, school teaching, etc., giving consumers a better user experience. The device can identify different scenes according to different sounds, thereby releasing different smells, allowing consumers to have a better experience, especially for game enthusiasts. During the game, they can also experience various smells such as car exhaust, gun smoke, blood, etc., so that the game experience is better. The device enables consumers to enjoy professional-level experience in various scenes at home. Through the device, the sounds in the game or video are collected, identified as various scenes, and then the corresponding smells are released according to different scenes. For example: in a racing game, the roar of the racing engine is identified as a racing scene. Through the device, the smell of the racing exhaust is automatically released, making people feel as if they are in a racing game, and getting a better game experience. For another example, when the sound of rain or leaves outside the window is identified, the fresh fragrance of the forest is released, making people feel like they are in nature. Because the sound is converted into smell, the olfactory experience is enhanced, adding a whole new experience to the audience watching movies, TV and playing games, in addition to vision and hearing, the sense of smell is also added.

[0046] The following is combined with Figure 1-6 , a detailed description of the automatic sound conversion device of an embodiment of the present invention is given.

[0047] like Figure 1As shown, the automatic sound conversion device according to an embodiment of the present invention includes an adapter, a server and an atomizer connected in sequence; wherein the adapter is used to collect sound signals and send the sound signals to the server; the server is used to identify the scene according to the sound signal, and map the scene to a corresponding odor code, and send the odor code to the atomizer; the atomizer is used to spray the corresponding odor according to the odor code.

[0048] Specifically, the source of the sound signal can be: game consoles, televisions, tablets and other home devices, which are input into the adapter through interfaces such as audioin / spdif in / hdmiin / ARC(EARC)in. The adapter will process the collected sound signal, and then use the RTSP protocol to transmit the sound in G.711 format to the server through WIFI (other common communication methods can also be used); the server will process the sound in real time, and then identify the real-time sound clip data as various scenes through the SED (sound event monitoring) model, and then map the scene to the corresponding odor code. The server sends the odor coding instruction to the sprayer, so that the sprayer sprays the odor of the corresponding scene.

[0049] It should be noted that a variety of essential oils with different scents can be pre-installed in the sprayer. When the sprayer receives a command from the server, it can emit the scent corresponding to the command.

[0050] According to the automatic sound conversion device of the embodiment of the present invention, different essential oils can be selected according to different scenes and different needs, so as to produce different smells. The device has a simple structure and is easy to carry. It can be widely used in home entertainment, Internet cafes, game halls, cinemas, school teaching, etc., giving consumers a better user experience. The device can identify different scenes according to different sounds, thereby releasing different smells, so that consumers have a better experience, especially for game enthusiasts. During the game, they can also experience various smells such as car exhaust, gun smoke, blood, etc., so that the game experience is better. The device enables consumers to enjoy professional-level experience in various scenes at home. Through the device, the sounds in the game or video are collected, identified as various scenes, and then the corresponding smells are released according to different scenes. For example: in a racing game, the roar of the racing engine is identified as a racing scene. Through the device, the smell of the racing exhaust is automatically released, making people feel as if they are in a racing game and get a better game experience. For another example, when the sound of rain or leaves outside the window is identified, the fresh fragrance of the forest is released, making people feel like they are in nature. Because the sound is converted into smell, the olfactory experience is enhanced, adding a whole new experience to the audience watching movies, TV and playing games, in addition to vision and hearing, the sense of smell is also added.

[0051] Furthermore, in some embodiments of the present invention, the adapter includes multiple sound transmission paths, a sound switching module, and a sound output module. The multiple sound transmission paths are used to receive sound signals from different interfaces. The sound switching module is used to select sound signals from the multiple sound transmission paths and send the selected sound signals to the server through the sound output module. By setting multiple sound transmission paths, the adapter can receive sounds from multiple different sources, thereby improving the applicability of the device and expanding the use scenarios of the device.

[0052] like Figure 2 As shown, one of the sound transmission paths (i.e., sound separation path 1) is shown, and the sound transmission path includes an optical fiber input interface (i.e., J9, SPDIF interface IN), a signal receiver (i.e., U18, signal receiver MS9113S) and an audio decoding chip (i.e., U20, MS8416T audio decoding); wherein, the optical fiber input interface J9 is used to receive a first sound signal (SPDIF_IN1), and the signal receiver U18 is used to amplify the first sound signal; the audio decoding chip U20 is used to decode the amplified first sound signal, obtain a first I2S signal (i.e., ①I2S), and send the first I2S signal to the sound switching module.

[0053] The above-mentioned sound path 1 is: the first sound signal received by the optical fiber input interface J9 passes through the signal receiver chip U18, amplifies the first sound signal to the power supply voltage 3.3V, and sends it to the audio decoding chip U20 to obtain the first I2S signal in the pcm format.

[0054] like Figure 3 As shown, another sound transmission path (i.e., sound separation path 2) is shown, which includes an HDMI input interface (i.e., CN1, HDMI interface IN) and an HDMI processing chip (i.e., U10, GSV2702 HDMI chip); wherein, the HDMI input interface CN1 is used to receive a first audio and video signal, and the HDMI processing chip U10 is used to extract audio information of the first audio and video signal, and output a second I2S signal (i.e., ②I2S) to the sound switching module.

[0055] The above-mentioned sound path 2 is: the HDMI input interface CN1 receives the audio and video signal, extracts the audio information in the HDMI signal through the HDMI processing core U10, and then outputs it through the I2S interface of the chip to obtain the second I2S signal.

[0056] It should be noted that if Figure 2 and Figure 3As shown, the optical fiber input interface J9 and the HDMI processing chip U10 are both connected to a first switching switch (i.e., U21, CH443K switch), which switches and selects the SPDIF_IN1 signal output by the optical fiber input interface J9 and the GSV_SPDIF_OUT signal output by the HDMI processing chip U10, and sends the selected signal (SPDIF_OUT) to the optical fiber output interface (i.e., J10, SPDIF interface OUT).

[0057] like Figure 4 As shown, one of the sound transmission paths (sound separation path 3) is shown, and the sound transmission path includes an HDMI output interface (i.e. CN3, HDMI interface OUT). The HDMI output interface CN3 is used to receive the second audio and video signal, and send the second audio and video signal to the HDMI processing chip U10, so that the HDMI processing chip U10 outputs the third I2S signal (i.e. ③I2S) to the sound switching module. The HDMI output interface CN3 is also connected to the first switching switch U21 through the HDMI processing chip U10, and then connected to the optical fiber output interface J10 through the first switching switch U21.

[0058] Specifically, Figure 4 As shown in the figure, the signal of HDMI output interface CN3 comes from HDMI OUT (ARC / EARC) of the TV, and the input audio signal will pass through HDMI chip U10 and the first switch U21, and finally output to SPDIF interface fiber output interface J10. At the same time, the third I2S signal will also be output through HDMI chip U10. Through this ARC (EARC) technology, high-quality sound can continue to be transmitted to the user's audio system.

[0059] like Figure 5 As shown, one of the sound transmission paths (sound separation path 4) is shown, which includes a headphone input interface (i.e. J22, headphone port IN) and an analog-to-digital conversion chip (i.e. U100, CJC5340 analog-to-digital conversion, LINE to I2S); wherein, the headphone input interface J22 is used to receive a second sound signal, and the analog-to-digital conversion chip U100 is used to convert the second sound signal into a pcm format, and output a fourth I2S signal (i.e. ④I2S) to the sound switching module.

[0060] Specifically, the left and right channel sounds input by the headphone input interface J22 are capacitively coupled to the sound analog-to-digital conversion chip U100, and the input sound signal is converted into a pcm format and output through the I2S interface of the chip to obtain the 4th I2S signal.

[0061] Furthermore, if Figure 5As shown, in some embodiments of the present invention, the sound switching module includes a second switching switch (i.e., U46 and U4, CH440G and RS2254 switching switches), a sound shunt unit (i.e., resistors PR1 and PR2), a digital-to-analog conversion chip (i.e., U101, CJC4344 digital-to-analog conversion, I2S to LINE) and a third switching switch (i.e., U17 / U5, CH442E / RS2227XUTQK10 switch); wherein, the second switching switch is used to switch the sound signals (①I2S, ②I2S, ③I2S, ④I2S) output by multiple sound transmission paths, select the corresponding sound signal, and output it to the sound shunt unit, the sound shunt unit is used to divide the sound signal into two paths, and send one of the sound signals to the server through the sound output module; the digital-to-analog conversion chip is used to perform digital-to-analog conversion on the other sound signal to obtain a third sound signal, the third switching switch is used to select the second sound signal and the third sound signal, and output the selected signal to the headphone output interface.

[0062] Specifically, in this example, the sound output module uses a WIFI module, and of course other common wireless communication modules can also be used. The 4-way I2S signal determines the final signal path through the second switch (U46 and U4), selects one of the I2S sounds, and divides it into two paths through the resistor (PR1, PR2), one path is uploaded to the server by the WIFI chip, and the other path is processed by the digital-to-analog conversion chip U101, and re-outputs the left and right channel sounds, and the signal input from the headphone input interface J22 before, and then flows to the third switch (U17 or U5) to decide which signal is finally output to the headphone output interface (i.e. J3, headphone port OUT).

[0063] According to the adapter of the embodiment of the present invention, after converting the input sound signal into pcm data through interfaces such as HDMI, headphone interface, and optical fiber interface, the data is transmitted to the WIFI module through the I2S bus, and then the WIFI module transmits the audio data to the server for further analysis and processing. By setting multiple sound transmission paths, the adapter can receive sound signals from multiple different sources, thereby improving the applicability of the device and making the application scenarios of the device wider.

[0064] Furthermore, in some embodiments of the present invention, the server includes: Wowza streaming server, Kafka, Lambda, SageMaker and a business server, wherein the adapter sends audio stream data to the Wowza streaming server according to the sound signal, and the Wowza streaming server sends the audio stream data to Kafka through the message queue; Lambda obtains the audio stream data by subscribing to Kafka messages, and adjusts the window size of the audio stream data to a preset window; SageMaker receives the audio stream data of the preset window, analyzes and infers events, obtains recognition results of the audio stream data, and sends the recognition results to Lambda, and Lambda sends the recognition results to Kafka's message queue; the business server is used to output the corresponding odor code to the sprayer according to the recognition results of Kafka's message queue.

[0065] Specifically, after the adapter sends the sound data to the cloud server, the main process is as follows:

[0066] (1) Audio stream reception

[0067] The audio data collected by the adapter is encoded into G711 format and sent to the Wowza streaming server in real time. The Wowza streaming server connects to the audio stream data through a custom plug-in, processes it, and forwards it to Kafka through a message queue.

[0068] (2) Message middleware forwarding

[0069] Lambda obtains real-time sound data by subscribing to Kafka messages, organizes it into data of the window size required by the model, and sends it to SageMaker.

[0070] (3) Model Reasoning

[0071] The sound event detection system based on the Transformer SED model is deployed on the AWS SageMaker platform. The audio stream data received by Lambda is transmitted to the SageMaker model through a pipeline. The model analyzes the sound data and performs event inference. The inference result includes the type and confidence of the recognized sound event.

[0072] (4) Reasoning result processing

[0073] The AWS Lambda function receives the inference results from SageMaker and sends them back to the inference result topic (inference_results) of the Kafka message queue. At this point, the business server subscribes to the inference result topic of Kafka and is ready to take corresponding actions based on the inference results.

[0074] (5) Business logic execution

[0075] The business server is responsible for the core management and control tasks. It will decide whether to send a spray instruction to the sprayer based on the inference results, such as triggering the release of odor when a specific sound event occurs. At the same time, the business server is also responsible for communicating with AWS IoT Core to perform device management (such as management of adapters and sprayers), user binding, odor capacity status monitoring and other functions.

[0076] (6) Equipment communication and control

[0077] AWS IoT Core communicates with the adapter and sprayer through the MQTT protocol, and sends the instructions issued by the business server (such as the spray instructions issued based on the inference results) to the corresponding devices to achieve control and status management of the terminal devices.

[0078] In some embodiments of the present invention, SageMaker's SED model uses the Transformer architecture, which is a non-convolutional, pure attention model that is directly applied to the audio spectrogram and can capture long-distance global context at the lowest level. Its main structure is as follows:

[0079] (1) Input processing: First, the model receives audio data and converts it into a two-dimensional audio spectrogram. This is usually done by converting the audio signal from the time domain to the frequency domain through short-time Fourier transform (STFT) to obtain the spectrogram as the input feature.

[0080] (2) Patch Embedding: Similar to the visual Transformer, the spectrogram is divided into multiple small patches of fixed size (patches), and then these patches are converted into a series of vector representations through a linear embedding layer, where each vector corresponds to a local area of ​​the spectrogram. This step is similar to the process of converting a two-dimensional image into a vector.

[0081] (3) Positional encoding: In order to preserve the temporal and spatial information of the audio spectrogram, the model adds positional encoding to the input patch embedding vector. This enables the model to identify the temporal relationship of the input data, thereby effectively understanding the time-varying characteristics of the audio.

[0082] (4) Transformer encoder: The core of the model is multiple stacked Transformer encoder layers, each of which contains a multi-head self-attention mechanism and a feedforward neural network. The self-attention mechanism can capture the global dependencies between different regions in the input spectrogram. Through this mechanism, the model learns the complex time-frequency features in the audio signal, thereby achieving accurate detection of sound events.

[0083] (5) Classification head: After being processed by the multi-layer Transformer encoder, the obtained global features are sent to a classification head. The classification head is usually composed of a linear layer, which is responsible for mapping the output of the Transformer to different sound event categories and generating the final classification result.

[0084] (6) Output: The final output is the probability distribution of the category to which each audio clip belongs. The model can determine the type of sound event it contains based on the characteristics of the input audio.

[0085] like Figure 6 As shown, in this example, the sprayer includes 6 atomization drivers WH1-WH6 (the number of atomization drivers can also be adjusted according to actual needs). Each atomization driver can drive the sprayer to spray different smells. When the sprayer is plugged into TYPE-C and powered on, the voltage is reduced to 3.3V through the LDO chip, which provides power for the chips in the system so that the system can work normally. The information sent by the server is received through the WIFI module (i.e. U2, WIFI 6220N), and the control command is parsed. The sprayer is driven by the control processor (i.e. U3 processor: SC8F2892B) so that a certain sprayer works and releases a specific smell. At the same time, the sprayer is also equipped with corresponding indicator lights / lighting lamps for each different smell.

[0086] According to the automatic sound conversion device of the embodiment of the present invention, different essential oils can be selected according to different scenes and different needs, so as to produce different smells. The device has a simple structure and is easy to carry. It can be widely used in home entertainment, Internet cafes, game halls, cinemas, school teaching, etc., giving consumers a better user experience. The device can identify different scenes according to different sounds, thereby releasing different smells, so that consumers have a better experience, especially for game enthusiasts. During the game, they can also experience various smells such as car exhaust, gun smoke, blood, etc., so that the game experience is better. The device enables consumers to enjoy professional-level experience in various scenes at home. Through the device, the sounds in the game or video are collected, identified as various scenes, and then the corresponding smells are released according to different scenes. For example: in a racing game, the roar of the racing engine is identified as a racing scene. Through the device, the smell of the racing exhaust is automatically released, making people feel as if they are in a racing game and get a better game experience. For another example, when the sound of rain or leaves outside the window is identified, the fresh fragrance of the forest is released, making people feel like they are in nature. Because the sound is converted into smell, the olfactory experience is enhanced, adding a whole new experience to the audience watching movies, TV and playing games, in addition to vision and hearing, the sense of smell is also added.

[0087] The embodiments of the present invention are described in detail above with reference to the accompanying drawings, but the present invention is not limited to the above embodiments, and various changes can be made within the knowledge scope of ordinary technicians in the relevant technical field without departing from the purpose of the present invention.

Claims

1. A sound automatic conversion device, characterized in that: include: An adapter, a server and a sprayer connected in sequence; wherein the adapter is used to collect sound signals and send the sound signals to the server; The server is used to identify the scene according to the sound signal, map the scene to a corresponding odor code, and send the odor code to the sprayer; the sprayer is used to spray the corresponding odor according to the odor code.

2. The automatic sound conversion device according to claim 1, characterized in that: The adapter includes multiple sound transmission paths, a sound switching module and a sound output module. The multiple sound transmission paths are used to receive sound signals from different interfaces. The sound switching module is used to select the sound signals from the multiple sound transmission paths and send the selected sound signals to the server through the sound output module.

3. The automatic sound conversion device according to claim 2, characterized in that: One of the sound transmission paths includes: An optical fiber input interface, used for receiving a first sound signal; A signal receiver, configured to amplify the first sound signal; The audio decoding chip is used to decode the amplified first sound signal to obtain a first I2S signal, and send the first I2S signal to the sound switching module.

4. The automatic sound conversion device according to claim 3, characterized in that: One of the sound transmission paths includes: An HDMI input interface, used for receiving a first audio and video signal; The HDMI processing chip is used to extract the audio information of the first audio and video signal and output a second I2S signal to the sound switching module.

5. The automatic sound conversion device according to claim 4, characterized in that: The HDMI processing chip is also connected to the optical fiber output interface through a first switching switch, and the optical fiber input interface is also connected to the optical fiber output interface through the first switching switch.

6. The automatic sound conversion device according to claim 5, characterized in that: One of the sound transmission paths includes: An HDMI output interface is used to receive a second audio and video signal, and send the second audio and video signal to the HDMI processing chip, so that the HDMI processing chip outputs a third I2S signal to the sound switching module. The HDMI output interface is also connected to the first switching switch through the HDMI processing chip, and then connected to the optical fiber output interface through the first switching switch.

7. The automatic sound conversion device according to claim 4, characterized in that: One of the sound transmission paths includes: An earphone input interface, used for receiving a second sound signal; The analog-to-digital conversion chip is used to convert the second sound signal into a pcm format and output a fourth I2S signal to the sound switching module.

8. The automatic sound conversion device according to claim 7, characterized in that: The sound switching module comprises: A second switching switch is used to switch the sound signals output by the plurality of sound transmission paths and select a corresponding sound signal; A sound splitting unit, used for dividing the sound signal into two paths, and sending one of the sound signals to the server through the sound output module; A digital-to-analog conversion chip, used for performing digital-to-analog conversion on another channel of the sound signal to obtain a third sound signal; The third switch is used to select between the second sound signal and the third sound signal, and output the selected signal to the headphone output interface.

9. The automatic sound conversion device according to claim 1, characterized in that: The server comprises: Wowza streaming media server, the adapter sends audio stream data to the Wowza streaming media server according to the sound signal; Kafka, the Wowza streaming server sends the audio stream data to the Kafka through a message queue; Lambda, obtains the audio stream data by subscribing to the Kafka message, and adjusts the window size of the audio stream data to a preset window; SageMaker receives the audio stream data of the preset window, performs analysis and event inference, obtains a recognition result of the audio stream data, and sends the recognition result to Lambda, and Lambda sends the recognition result to the message queue of Kafka; The business server is used to output the corresponding odor code to the sprayer according to the recognition result of the Kafka message queue.

10. The automatic sound conversion device according to claim 1, characterized in that: The business server also communicates with AWS IoT Core, and manages and controls the adapter and the sprayer through the AWS IoT Core.

Images