WO2020181407A1

WO2020181407A1 - Voice recognition control method and device

Info

Publication number: WO2020181407A1
Application number: PCT/CN2019/077469
Authority: WO
Inventors: 陈旻宏
Original assignee: 发条橘子云端行销股份有限公司
Priority date: 2019-03-08
Filing date: 2019-03-08
Publication date: 2020-09-17

Abstract

A voice recognition control method and a voice recognition control device (1). The voice recognition control method comprises: receiving a voice signal with a voice transceiver (20) (S1); learning about the voice signal with a learning module (40) interacting at least with a cloud search engine server (3), then converting the voice signal into at least one piece of voice data (S2); parsing the voice signal with a processor (50) on the basis of language data to generate a control signal and a voice feedback signal (S3); and transmitting the control signal with an infrared transceiver (10) to control at least one piece of electrical equipment (2) (S4). The voice recognition control device (1) comprises the voice transceiver (20), the learning module (40), the processor (50), a storage unit (30), and the infrared transmitter (10). Voice control of the at least one piece of electrical equipment (2) is implemented via the voice recognition control method and the voice recognition control device (1).

Description

Voice recognition control method and device

Technical field

This application is related to voice control, especially a voice recognition control method and device.

Background technique

With the advancement of science and technology, home equipment is gradually leading to the concept of smart home, using automated systems to adjust the home environment, and to improve the problem of only one remote control for specific electrical equipment in the past.

The related art involves a controller controlling a plurality of electrical devices, and the controller is usually a smart phone, a tablet computer or other remote controllers, which controls the operating state of the multiple electrical devices in a wired or wireless manner. However, the controller of the related art must be operated by hand to achieve the control effect. Such a complicated operation method is really inconvenient for the elderly or other users.

Therefore, it is necessary to provide a voice recognition control method and device to solve the above-mentioned problems.

Summary of the invention

The main purpose of this application is to provide a voice recognition control method and device to control at least one electrical device by voice.

In order to achieve the above objective, this application provides a voice recognition control method, including: receiving a voice signal with a voice transceiver; using a learning module to interact with at least a cloud search engine server to learn the voice signal, and then converting the voice signal Is at least one language data; a processor analyzes the voice signal according to each of the language data to generate a control signal; and an infrared transmitter transmits the control signal for controlling at least one electrical device.

In order to achieve the above-mentioned object, the present application further provides a voice recognition control device, including: an infrared transmitter for transmitting at least one control signal, the at least one control signal for controlling at least one electrical device; a voice transceiver, which Used for receiving a voice signal and transmitting a voice feedback signal generated according to the voice signal; a learning module that at least converts the voice signal into at least one language data; a storage unit that stores each of the language data and at least one environment At least one of data and at least one state data; and a processor for analyzing the voice signal according to each of the language data, and reading the at least one environmental data and the at least one state data to convert the voice signal into the control At least one of the signal and the voice feedback signal.

Optionally, the learning module is connected to a cloud search engine server, and the learning module is used to obtain the at least one language data from the cloud search engine server according to the voice signal.

Optionally, the processor further includes a voice recognition unit and a semantic recognition unit, the voice recognition unit analyzes the voice signal as at least one text signal, and the semantic recognition unit determines each text signal according to the at least one language data, the at least one At least one of an environmental data and the at least one state data is converted into at least one of the control signal or the voice feedback signal.

Optionally, the storage unit further includes a semantic database, and the semantic database stores each of the language data.

Optionally, the storage unit further includes an environment database and a state database, the environment database stores each of the environmental data, and the state database stores each of the state data.

Optionally, the voice transceiver further includes an echo filtering module, and when the voice signal contains an echo signal, the echo filtering module filters the echo signal.

Optionally, the voice transceiver further includes a noise filtering module, and when the voice signal includes a noise signal, the noise filtering module can filter the noise signal.

Optionally, the voice transceiver includes a radio module and a sound module, the radio module receives the voice signal, and the sound module transmits the voice feedback signal.

Optionally, the number of the at least one language data, the at least one environmental data, and the at least one state data is multiple; the learning module is connected to a cloud search engine server, and the learning module is used to send the voice signal to the The cloud search engine server obtains the at least one language data; the storage unit further includes a semantic database that stores each of the language data; the storage unit further includes an environment database and a state database, the environment database stores each of the environmental data , The state database stores the state data; the voice transceiver further includes an echo filtering module, when the voice signal contains an echo signal, the echo filtering module filters the echo signal; the voice transceiver further includes an echo signal The noise filter module, when the voice signal includes a noise signal, the noise filter module filters the noise signal; the voice transceiver includes a radio module and a sound module, the radio module receives the voice signal, the speaker The voice module transmits the voice feedback signal; the voice transceiver further includes a playback module connected to the semantic recognition unit and the playback module; the voice recognition control device further includes a display connected to the processor, the The display can display at least one image information.

Description of the drawings

Fig. 1 is a flowchart of a preferred embodiment of this application.

Figure 2 is a block diagram of a preferred embodiment of the application.

FIG. 3 is a schematic diagram of a use state of a preferred embodiment of the application.

Symbol description: S1 to S4: steps; 1: voice recognition control device; 2: electrical equipment; 3: cloud search engine server; 4: user; 10: infrared transmitter; 20: voice transceiver; 21: echo filter Module; 22: Noise Filter Module; 23: Radio Module; 24: Playback Module; 25: Play Module; 30: Storage Unit; 31: Semantic Database; 32: Environmental Database; 33: State Database; 40: Learning Module; 50: processor; 51: speech recognition unit; 52: semantic recognition unit.

detailed description

In order to make the purpose, technical solutions, and advantages of this application clearer, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the application, and not used to limit the application. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative work shall fall within the protection scope of this application.

Please refer to FIG. 1, which shows a preferred embodiment of the present application. The voice recognition control method of the present application includes the following steps: Step S1: Receive a voice signal with a voice transceiver; Step S2: Use a learning module to at least interact with A cloud search engine server interactively learns the voice signal, and then converts the voice signal into at least one language data. It is further explained that the learning module can be connected to the cloud search engine server in a wired or wireless manner to search for the semantics and grammar of the voice signal. Language data; Step S3: Use a processor to analyze the voice signal according to each of the language data to generate a control signal and a voice feedback signal; and Step S4: Use an infrared transmitter to transmit the control signal to control at least one electrical equipment. In addition, the learning module can also use the voice feedback signal to answer the user's questions or issue questions, and can learn the semantics and grammar of the language signal through interaction with the user.

2 to 3, the voice recognition control device 1 of the present application includes an infrared transmitter 10, a voice transceiver 20, a storage unit 30, a learning module 40, and a processor 50.

The infrared transmitter 10 can emit at least one control signal, each of the control signals is an infrared signal, and the at least one control signal is used to control at least one electrical device 2. In this embodiment, the at least one electrical device 2 can receive the infrared signal Of course, the air conditioner, lamp, TV, or fan can also be an electrical device 2 that sends infrared signals; in other embodiments, the infrared transmitter 10 can also emit multiple control signals to control multiple electrical devices at the same time.

The voice transceiver 20 can receive a voice signal and can transmit a voice feedback signal generated according to the voice signal; the learning module 40 at least converts the voice signal into at least one language data; the storage unit 30 stores each of the language data, At least one of at least one environmental data and at least one state data; and the processor 50 parses the voice signal according to each of the language data, and reads at least one of the at least one environmental data and the at least one state data to perform the The voice signal is converted into at least one of the control signal and the voice feedback signal. Thereby, the voice recognition control device 1 can interact with the user 4 to learn Chinese grammar, and can control each of the electrical equipment 2 through the voice signal, so as to improve the convenience of operation.

In this embodiment, the number of the at least one language data, the at least one environment data, and the at least one status data is multiple respectively; the multiple language data may include Chinese, English, Cantonese, Hokkien, Thai and other languages. Vocabulary and grammar; the multiple environmental data may include multiple environmental names, and the multiple state data may include environmental temperature status, environmental humidity status, operating status of the at least one electrical device 2 and so on.

Preferably, the learning module 40 can be connected to a cloud search engine server 3, which can be connected to the cloud search engine server 3 in a wired or wireless manner. The cloud search engine server 3 can be network information such as a search engine (such as GOOGLE), an information database (such as Wikipedia), and the learning module 40 can obtain the at least one language data from the cloud search engine server 3 according to the voice signal, and so The learning module 40 can learn through the network.

In detail, the processor 50 further includes a voice recognition unit 51 and a semantic recognition unit 52. The voice recognition unit 51 analyzes the voice signal as at least one text signal, and the semantic recognition unit 52 converts each text signal according to the at least one text signal. At least one of the language data, the at least one environmental data, and the at least one state data is converted into at least one of the control signal or the voice feedback signal, so that the voice signal can be clearly analyzed and interpreted. In addition, the voice recognition unit 51 can determine different pronunciations and intonations to match similar characters.

The storage unit 30 further includes a semantic database 31, and the semantic database 31 stores each language data. In addition, the storage unit 30 further includes an environmental database 32, and the environmental database 32 stores various environmental data. Further, the storage unit 30 further includes a state database 33, and the state database 33 stores each state data.

The voice transceiver 20 further includes an echo filter module 21. When the voice signal contains an echo signal, the echo filter module 21 can filter the echo signal. In addition, the voice transceiver 20 further includes a noise filter module 22. When the voice signal includes a noise signal, the noise filter module 22 can filter the noise signal to improve the clarity of the voice signal.

The voice transceiver 20 includes a receiving module 23 and a playing module 24. The receiving module 23 can receive the voice signal, and the playing module 24 can transmit the voice feedback signal. The sound receiving module 23 may be, for example, a microphone device; the sound playback module 24 may be, for example, a speaker device. In addition, the voice transceiver 20 further includes a playing module 25 connected to the semantic recognition unit 52 and the sound playing module 24.

In this embodiment, the voice recognition control device 1 further includes a display (not shown in the figure), the display is connected to the processor 50, the display can display at least one image information, the at least one image information can be multimedia , Or remote video image to interact with the user 4 with images.

The above are only preferred embodiments of this application, and do not limit the scope of this application. Any equivalent structure or equivalent process transformation made using the content of the description and drawings of this application, or directly or indirectly used in other related technical fields , The same reason is included in the scope of patent protection of this application.

Industrial applicability

The voice recognition control method and device provided by the embodiments of the present application uses a voice transceiver to receive a voice signal; uses a learning module to interact with at least a cloud search engine server to learn the voice signal, and then convert the voice signal to at least A language data; a processor analyzes the voice signal according to each of the language data to generate a control signal and a voice feedback signal; and an infrared transmitter transmits the control signal for controlling at least one electrical device. At least one electrical device is realized by sound control, so it has industrial applicability.

Claims

A voice recognition control method, which includes:

Receive a voice signal with a voice transceiver;

Use a learning module to interact with at least one cloud search engine server to learn the voice signal, and then convert the voice signal into at least one language data;

Using a processor to analyze the voice signal according to each of the language data to generate a control signal and a voice feedback signal; and

An infrared transmitter is used to transmit the control signal for controlling at least one electrical device.
A voice recognition control device, which includes:

An infrared transmitter for emitting at least one control signal for controlling at least one electrical device;

A voice transceiver for receiving a voice signal and transmitting a voice feedback signal generated according to the voice signal;

A learning module, at least converting the voice signal into at least one language data;

A storage unit storing at least one of each of the language data, at least one environmental data, and at least one state data; and

A processor analyzes the voice signal according to each of the language data, and reads the at least one environmental data and the at least one state data to convert the voice signal into at least one of the control signal and the voice feedback signal.
3. The voice recognition control device of claim 2, wherein the learning module is connected to a cloud search engine server, and the learning module is used to obtain the at least one language data from the cloud search engine server according to the voice signal.
3. The voice recognition control device of claim 2, wherein the processor further comprises a voice recognition unit and a semantic recognition unit, the voice recognition unit analyzes the voice signal as at least one text signal, and the semantic recognition unit converts each of the text The signal is converted into at least one of the control signal or the voice feedback signal according to at least one of the at least one language data, the at least one environmental data, and the at least one state data.
3. The voice recognition control device of claim 2, wherein the storage unit further comprises a semantic database, the semantic database storing each of the language data.
3. The voice recognition control device of claim 2, wherein the storage unit further comprises an environment database and a state database, the environment database stores each of the environmental data, and the state database stores each of the state data.
3. The voice recognition control device of claim 2, wherein the voice transceiver further comprises an echo filtering module, and when the voice signal includes an echo signal, the echo filtering module filters the echo signal.
3. The voice recognition control device of claim 2, wherein the voice transceiver further comprises a noise filtering module, and when the voice signal includes a noise signal, the noise filtering module can filter the noise signal.
3. The voice recognition control device of claim 2, wherein the voice transceiver comprises a radio module and a sound module, the radio module receives the voice signal, and the sound module transmits the voice feedback signal.
4. The voice recognition control device of claim 4, wherein the number of the at least one language data, the at least one environment data, and the at least one status data is multiple; the learning module is connected to a cloud search engine server, the The learning module is used to obtain the at least one language data from the cloud search engine server according to the voice signal; the storage unit further includes a semantic database, the semantic database stores each language data; the storage unit further includes an environment database and a status The environment database stores each of the environmental data, the state database stores each of the state data; the voice transceiver further includes an echo filter module, when the voice signal contains an echo signal, the echo filter module filters the Echo signal; the voice transceiver further includes a noise filter module, when the voice signal includes a noise signal, the noise filter module filters the noise signal; the voice transceiver includes a radio module and a playback module, The radio module receives the voice signal, and the playback module transmits the voice feedback signal; the voice transceiver further includes a playback module connected to the semantic recognition unit and the playback module; the voice recognition control device further includes a The display is connected to the processor, and the display can display at least one image information.