TW202230245A - Voice marketing information system and method - Google Patents

Abstract

The invention discloses a voice marketing information system and method, including: a speaker device, including: an audio receiving device for receiving a user's voice signal; and an audio emitting device; a semantic recognition feedback module, including: an audio conversion A unit that converts the user's voice signal into a text data; an order structure data unit that compares the text data with the order structure data unit to generate one order data; and a customer service feedback unit; and a sales The point-in-time transaction module includes: a prompt unit, a form generating unit, and a customer service information unit. The voice marketing information system and method of the present invention can achieve the effect of voice unmanned order.

Description

Voice marketing information system and method

The present invention relates to a marketing information system and method, in particular to a voice marketing information system and method.

Commonly known beverage stores, snack bars, restaurants, service industries and retail industries, etc., usually require a service person to use the marketing POS system (Point of sale) to carry out the marketing process to reduce errors in the service process and provide sales. In addition to increasing personnel costs and wasting manpower, this method also increases the contact between personnel and increases the risk of infecting each other with various diseases. In recent years, although there have been Self-service ordering machines are operated by consumers to select service items by touching the screen. At present, the novel coronavirus epidemic is serious, and operating self-service ordering machines will increase the risk of indirect contact infection, which will bring new challenges to the epidemic prevention policy. The machine is operated by consumers themselves, and the screen is used as the interface to allow consumers to choose, which requires consumers to be familiar with different operating interfaces. It may happen that consumers touch the screen by mistake or the operation process is complicated, which makes the consumer operation process lengthy and large queues appear Long-term inefficient consumption behavior increases time and labor costs in the process of consumption behavior, and also makes it difficult for visually impaired or illiterate people to use the service, and a complete ordering machine, in addition to the problem of occupying space, its price is higher. It is expensive and not all businesses can afford it.

Although people can use the phone to order food by voice, the consumption behavior may not be determined in advance, and they often walk into the store on a temporary basis. In addition, some people with language barriers, such as sheltered workshops, sheltered stores or general caring stores, are often unable to order food or communicate with customers.

The purpose of the present invention is to solve the problem that the prior art marketing POS system wastes manpower and increases infectious diseases due to contact between people. Another purpose of the present invention is to solve the problem that the introduction of unmanned ordering machines on the market is expensive and ordering machines In order to achieve the above-mentioned purpose, the present invention provides a marketing information system, including: a speaker device, including: an audio receiving device, which receives a user voice signal sent by a user; and a The audio sending device sends out a customer service audio; a semantic recognition feedback module is connected with the speaker device information, including: an audio conversion unit, which converts the user's voice signal into a text data; an order structure data unit, The text data is compared with the order structure data unit to generate an order data; and a customer service feedback unit, which can convert a text customer service information into a customer service voice signal and transmit it to the audio signal issuing device; and a point-of-sale transaction The module, which is connected with the information of the semantic recognition feedback module, includes: a prompting unit for prompting that the order data has been received; a form generating unit for sequentially inputting the order data according to the order data and generating an order form; and a The customer service information unit can generate a text of customer service information and send it to the customer service feedback unit.

Wherein, the above-mentioned point-of-sale transaction module may be a cloud module or a physical POS machine.

In an embodiment of the present invention, the sound box device further includes a position information unit, wherein the sound box device is disposed at a table, and the position information unit can set the corresponding seat information of the sound box device for use Directly at the seat to make a tableside voice ordering.

Wherein, the speaker device further includes an infrared sensing device, which can sense the approach of the user and issue a welcome sound.

In an embodiment of the present invention, the prompt unit further has a bell unit and a confirmation interface. After the prompt unit receives the order data, the bell unit makes a sound and displays the confirmation interface for a The clerk confirms.

In an embodiment of the present invention, the point-of-sale transaction module further includes a checkout unit for the user to perform voice checkout.

In an embodiment of the present invention, the semantic recognition feedback module further includes an audio comparison module and an audio comparison database, wherein the audio comparison module can compare the audio comparison database with the audio comparison database. The frequency of the user's voice determines whether the user is a person with the ability to consume.

In an embodiment of the present invention, the semantic recognition feedback module further includes a debug unit, which is based on a preset value range ratio to the order form. When the value range is exceeded, the debug unit will issue a Error audio signal to the speaker device.

The present invention discloses a voice marketing method, comprising the following steps: sending a voice problem signal based on the field information required by the order configuration data unit; receiving a user voice signal; comparing the voice frequency of the user voice signal to determine the use Whether the user has the ability to consume behavior; perform semantic recognition, convert the user's voice signal into text data; compare the text data based on an order structure data unit and convert it into an order data; check whether the value of the order data is reasonable ; Generate an order form based on the order data; send out a sound to prompt the store to confirm the order form; send an order-accepted voice message to the user.

The marketing information system and method of the present invention can achieve the effect of unmanned voice automatic ordering, greatly reduce labor costs, and reduce direct contact between personnel, and the marketing information system and method provided by the present invention can be used through simple and cheap speakers. equipment, so it can be easily imported General merchants and stores can reduce the direct dialogue and contact between employees and customers, and even achieve the effect of contactless ordering. Another effect of the present invention is that the present invention can achieve simple and cheap speaker equipment, so that consumers can Being able to sit in a seat can directly order food by voice, avoiding unnecessary contact between various personnel, and saving the labor cost of merchants.

00: user

10: Speaker device

11: Audio receiving device

12: Audio sending device

13: Location Information Unit

14: Infrared induction device

20: Semantic recognition feedback module

21: Audio conversion unit

22: Order structure data unit

23: Customer Service Feedback Unit

30: Point-of-sale transaction module

31: Cue Unit

32: Form generation unit

33: Customer Service Information Unit

34: Checkout Unit

40: Bank side

S401~S409: Step flow

FIG. 1 is a schematic diagram of a voice marketing information system.

FIG. 2 is a flow chart of steps of an embodiment.

In order to enable those with ordinary knowledge in the technical field to understand the content of the present invention and realize the content of the present invention accordingly, the following description is given with appropriate embodiments in conjunction with the drawings, and equivalent replacements and modifications based on the content of the present invention are made. All are included in the scope of rights of the present invention.

Please refer to FIG. 1 , which is a schematic diagram of a voice marketing information system.

The voice marketing information system 1 according to the first embodiment of the present invention includes a speaker device 10 , a semantic recognition feedback module 20 and a point-of-sale transaction module 30 . The speaker device 10 is informationally connected to the semantic recognition feedback module 20 . , the semantic recognition feedback module 20 is connected with a point-of-sale transaction module 30 for information, and the information connection can be wired or wireless. In this embodiment, the information is connected by wireless network. connection, wherein the speaker device 10 has an audio receiving device 11 and an audio sending device 12, the semantic recognition feedback module 20 has an audio conversion unit 21, an order structure data unit 22 and a customer service feedback unit 23, the point-of-sale transaction model The group 30 has a prompting unit 31, a form generating unit 32 and a customer service information unit 33; the audio receiving device 11 receives a user voice signal sent by a user 00, and the speaker device 10 transmits the user voice signal information To the audio conversion unit 21, the audio conversion unit 21 converts the user's voice signal into a text data The order structure data unit 22 compares the text data with the order structure data unit 22 to generate an order data, the order data information is sent to the point-of-sale transaction module 30, and the prompt unit 31 of the point-of-sale transaction module 30 generates an order data, and a reminder that the order data has been received, the form generating unit 32 sequentially inputs the order data according to the order data, and generates an order form, and the customer service information unit 33 generates a textual customer service information based on the order form and sends it to the customer service feedback unit 23. The customer service feedback unit 23 converts the text customer service information into audio signal information and transmits it to the audio signal sending device 12, and emits a customer service voice.

Preferably, the audio conversion unit 21 converts the user's voice signal into text data by using STT (speech to text) technology; the customer service feedback unit 23 converts the text customer service information into audio signal information by using TTS (text to speech) technology. .

The point-of-sale transaction module 30 may be a cloud module or a point-of-sale information system POS (point of sale) machine, which is a point of sale (POS) machine in this embodiment.

The order structure data unit 22 has at least one preset order structure data. In this embodiment, the order structure data unit 22 has an item field, a quantity field, and a feature field, but it is not limited to this. The field, the quantity field, and a characteristic field are all provided with corresponding keyword data for the text data to be compared with the order structure data unit 22 .

In one embodiment, the item field database has an item name AAC, the default keywords are AAC, AC, Aac, aC, and ac, and the text data corresponds to the preset keywords in the item field database to All are converted into item name AAC. In this embodiment, the recognition speed of the semantic recognition feedback module 20 can be accelerated.

Preferably, when the text data is compared with the order structure data unit 22 and the corresponding data cannot be found, the customer service feedback unit 23 sends a voice signal of "I'm sorry, I don't understand clearly, please say it again" to the audio sending device 12 .

In another embodiment, the order structure data unit 22 has an item field, a quantity field, an ice degree field, and a sugar content field in sequence, and the item field has an item name: "pearl milk tea", Its default keywords are "Bubble Milk Tea", "Zhen Milk", "Zhen Milk Tea"; the ice degree field has an ice degree name "De-Ice", and its default keywords are "De-Ice", "No Ice" , "no ice"; the sugar content column has a sugar content name "micro sugar", and its default keywords are "micro sugar" and "three-point sugar"; the audio receiving device 11 receives a message sent by a user 00 to use The speaker device 10 transmits the user's voice signal information to the audio conversion unit 21, and the audio conversion unit 21 converts the user's voice signal into a text data "Boss, have a cup of precious milk". , the order structure data unit 22 compares the text data with the order structure data unit 22 to generate an order data, the item field is "bubble milk tea", and the quantity field is "one cup". For the column information and the sugar content column information, the customer service feedback unit 23 sends a voice signal “Ask the sugar content and ice cubes” to the audio sending device 12, and the audio receiving device 11 receives a user voice signal “Three-point sugar” sent by the user 00. "Remove ice", the speaker device 10 transmits the user's voice signal information to the audio conversion unit 21, and the audio conversion unit 21 converts the user's voice signal into a text data "Three points of sugar to remove the ice", the order structure data unit 22 The text The data is compared with the data unit 22 of the order structure, the column of ice degree is "De-ice", and the column of sugar content is "micro sugar", complete order data is generated, and the order data information is sent to the transaction module 30 at the point of sale, and the time of sale The prompting unit 31 of the transaction module 30 prompts that the order data has been received, and the form generating unit 32 inputs sequentially according to the order of the fields set by the order data, and generates an order form, wherein the order form further includes the price information of bubble tea, The customer service information unit 33 generates a text customer service information based on the order form and transmits it to the customer service feedback unit 23. The customer service feedback unit 23 converts the text customer service information into audio signal information and transmits it to the audio signal sending device 12, and sends out a customer service voice "Started. Make your order, a cup of bubble milk tea with ice and sugar, a total of 45 yuan, please wait".

In one embodiment, the sound box device 10 further includes a position information unit 13 , wherein the sound box device 10 is arranged on the edge of a table, and the position information unit 13 can set the corresponding position of the sound box device 10 . The seat information is for the user 00 to order at the table, and the seat information is transmitted to the order structure data unit 22 .

Preferably, a plurality of speaker devices 10 are disposed at a plurality of tables, and the plurality of speaker devices 10 have a plurality of different seat information for a plurality of users 00 to order food at the table at the same time.

Preferably, the speaker device 10 further includes an infrared sensing device 14, which can sense the approach of the user 00 and issue a welcome message. Not limited to this welcome audio content.

Preferably, the prompting unit 31 further has a bell unit and a confirmation interface. After the prompting unit 31 receives the order data, the bell unit makes a sound and displays the confirmation interface for the clerk to confirm.

In one embodiment, the point-of-sale transaction module 30 further includes a checkout unit 34, the checkout unit 34 is connected with the bank terminal 40 for information, for the user 00 to perform voice checkout, wherein the order structure data The unit 22 further has a credit card number field; the user 00 sends a voice message of the credit card number, the speaker device 10 transmits the user voice signal information to the audio conversion unit 21, and the audio conversion unit 21 converts the user voice signal into a text data "Credit card number", the order structure data unit 22 compares the text data with the order structure data unit 22 to generate an order data, the order data information is sent to the point-of-sale transaction module 30, and the prompt unit of the point-of-sale transaction module 30 31 generates an order data, and prompts that the order data has been received, the form generation unit 32 inputs sequentially according to the order data, and generates an order form, and the checkout unit 34 obtains the "credit card number" information in the order form, "credit card number" Card number” information is sent to the bank terminal 40 for checkout, the customer service information unit 33 sends the checkout information to the customer service feedback unit 23, and the customer service feedback unit 23 converts the text customer service information into audio signal information and transmits it to the audio output. device 12 .

Preferably, the semantic recognition feedback module 20 further includes an audio comparison module and an audio comparison database, wherein the audio comparison database has a plurality of voices of people who do not have the ability to consume behavior. Audio frequency data, among which, those who do not have the ability to consume refers to children under the age of seven as defined by the laws of our country. Whether the user 00 is a person with the ability to consume behavior, when the sound frequency of the user 00 is similar to the sound frequency data of the person who does not have the ability to consume behavior, the audio-emitting device 12 sends a sound signal "Are you over seven years old?", but Not limited to the content of the audio signal, when the user 00 answers yes, the ordering process is continued; otherwise, the ordering process is ended.

Preferably, the semantic recognition feedback module 20 further includes a debug unit, which is based on a preset value range compared to the order data. When the value range is exceeded, the debug unit will issue an error sound signal to the The audio output device 12 .

In one embodiment, the debug unit has a value range of 1 to 20. When a value in the order data is 100, the debug unit will send out an error sound signal "Are you ordering 100 cups?" to the audio sending device. 12, but not limited to the content of the error voice signal, if the user 00 answers yes, then continue the ordering process; otherwise, if the user 00 answers no, the numerical data will be deleted, and the audio sending device 12 will resend the question to obtain the correct value material.

Please refer to FIG. 2. FIG. 2 is a flow chart of the steps of an embodiment. The main steps are as follows: Step S401, based on the field information required by the order structure data unit 22, send a voice problem signal; Step S402, receive a user voice signal; step S403, comparing the sound frequency of the user's voice signal to determine whether the user 00 has the ability to consume behavior, the audio comparison module and an audio comparison database are used for frequency approximation judgment, the audio comparison The database has the voice frequency data of a plurality of people who do not have the ability to consume, among which, the people who do not have the ability to consume refers to children under the age of seven under the laws of our country; step S404, perform semantic recognition, and convert the user's voice signal into a text data; step S405, the text data is compared based on an order structure data unit 22 and converted into an order data, wherein the order structure data unit 22 has at least one preset order structure data, the order structure data The data is provided with at least one preset field and its corresponding keyword for comparison; step S406 , checking whether the value of the order data is reasonable is based on a preset value range; step S407 , based on the order data An order form is generated; in step S408, a sound is emitted and a confirmation interface pops up, prompting the store to confirm the order form 321; in step S409, an order received voice message is sent to the user 00.

00: user

10: Speaker device

11: Audio receiving device

12: Audio sending device

13: Location Information Unit

14: Infrared induction device

20: Semantic recognition feedback module

21: Audio conversion unit

22: Order structure data unit

23: Customer Service Feedback Unit

30: Point-of-sale transaction module

31: Cue Unit

32: Form generation unit

33: Customer Service Information Unit

34: Checkout Unit

40: Bank side

Claims (10)

A voice marketing information system, comprising: A loudspeaker device comprising: an audio receiving device for receiving a user voice signal sent by a user; and an audio message sending device, which sends out a customer service audio; A semantic recognition feedback module, which is connected with the speaker device information, includes: an audio conversion unit, which converts the user's voice signal into a text data; an order structure data unit that compares the text data to generate an order data; and a customer service feedback unit, which can convert a text customer service information into a customer service voice signal and transmit it to the audio signal sending device; and A point-of-sale transaction module, which is connected to the information of the semantic recognition feedback module, including: a prompting unit, indicating that the order information has been received; a form generating unit for generating an order form according to the order data; and A customer service information unit can generate a text customer service information and send it to the customer service feedback unit. The voice marketing information system according to claim 1, wherein the sound box device further comprises a position information unit, wherein the sound box device is arranged on the edge of a table, and the position information unit can set the corresponding seat information of the sound box device , for users to order at the table. The voice marketing information system according to claim 1, wherein the speaker device further comprises an infrared sensor device, which can sense the user's approach and issue a welcome message. The voice marketing information system according to claim 1, wherein the point-of-sale transaction module is a cloud module. The voice marketing information system according to claim 1, wherein the point-of-sale transaction module is a POS machine. The voice marketing information system according to claim 1, wherein the prompt unit further has a bell unit and a confirmation interface. After the prompt unit receives the order data, the bell unit makes a sound and displays The confirmation interface is for a store clerk to confirm. The voice marketing information system of claim 1, wherein the point-of-sale transaction module further comprises a checkout unit, which is connected to a bank-side information for the user to perform voice checkout. The voice marketing information system of claim 1, wherein the semantic recognition feedback module further comprises an audio comparison module and an audio comparison database, wherein the audio comparison module can compare the audio comparison Based on the database and the user's voice frequency, determine whether the user is a person with the ability to consume. The voice marketing information system according to any one of claims 1 to 8, wherein the semantic recognition feedback module further comprises a debug unit, based on a preset value range ratio to the order data, when the When the value is in the range, the debug unit will send out an error sound signal to the audio sending device. A voice marketing method that includes the following steps: Send an audio problem signal based on the field information required by the order structure data element; receive a user voice signal; Compare the sound frequency of the user's voice signal to determine whether the user has the ability to consume; Perform semantic recognition to convert the user's voice signal into a textual data; comparing the text data based on an order structure data unit and converting it into an order data; Check whether the value of the order information is reasonable; generating an order form based on the order data; sound an audible prompting the store to confirm the order form; and Send an order received voice message to the client.

Images