JP6807586B1 - Information processing equipment, information processing methods and programs - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a technique for improving the accuracy of voice recognition of voice data of business negotiations. In an information processing system, an information processing device (server device 100) has a voice recognition unit 401 and a registration unit 403. The voice recognition unit performs voice recognition based on the voice data of the negotiation, converts it into text, and recognizes the word based on the data of the word conversion dictionary if there is a word that cannot be recognized when performing voice recognition. The registration department registers a new word in the word conversion dictionary based on the registration request. [Selection diagram] Fig. 4

Description

The present invention relates to an information processing device, an information processing method and a program.

There is a method called inside sales that conducts sales activities remotely to customers.
Patent Document 1 discloses a promising customer prediction device that can also be used for inside sales.

Japanese Patent No. 6031165

However, you may want to record a video of your inside sales and convert it into text to see what the sales person was saying later. For example, when a boss wants to check how his subordinates are doing business, or when he wants to show a video of a person with good sales results to a new sales person as an education. In such cases, unusual company and service names may appear in sales conversations. Uncommon company names and service names are difficult to recognize correctly in speech recognition.

The present invention has been made in view of such circumstances, and an object of the present invention is to provide a technique for improving the accuracy of voice recognition of voice data of business negotiations.

According to one aspect of the present invention, the information processing device includes a voice recognition unit and a registration unit, and the voice recognition unit performs voice recognition based on the voice data of a business negotiation, converts it into a text, and describes the above. When performing voice recognition, if there is an unrecognizable word, the word is recognized based on the data of the word conversion dictionary, and the registration unit registers a new word in the word conversion dictionary based on the registration request. An information processing device is provided.

According to one of the present inventions, there is an advantageous effect that a technique for improving the accuracy of speech recognition can be provided.

FIG. 1 is a diagram showing an example of a system configuration of an information processing system. FIG. 2 is a diagram showing an example of the hardware configuration of the server device. FIG. 3 is a diagram showing an example of the hardware configuration of the client device. FIG. 4 is a diagram showing an example of the functional configuration of the server device. FIG. 5 is an activity diagram showing an example of information processing related to registration of the server device. FIG. 6 is an activity diagram showing an example of information processing related to search in the server device. FIG. 7 is a diagram showing an example of the search result screen. FIG. 8 is a diagram showing an example of a list screen.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. The various features shown in the embodiments shown below can be combined with each other.

As used herein, the term "part" may include, for example, a combination of hardware resources implemented by circuits in a broad sense and information processing of software that can be specifically realized by these hardware resources. Further, various information is handled in this embodiment, and these information are represented by high and low signal values as a bit set of binary numbers composed of 0 or 1, and communication / calculation is executed on a circuit in a broad sense. Can be done.

Further, a circuit in a broad sense is a circuit realized by at least appropriately combining a circuit (Circuit), circuits (Circuitry), a processor (Processor), a memory (Memory), and the like. That is, an integrated circuit for a specific application (Application Special Integrated Circuit: ASIC), a programmable logic device (for example, a simple programmable logic device (Simple Programmable Logical Device: SPLD), a composite programmable logic device (Complex Programg)). It includes a programmable gate array (Field Programmable Gate Array: FPGA) and the like.

<Embodiment 1>
1. 1. System Configuration FIG. 1 is a diagram showing an example of the system configuration of the information processing system 1000. As shown in FIG. 1, the information processing system 1000 includes a server device 100, a client device 110, a client device 120, and a client device 130 as a system configuration. The client device 110 is a personal computer (Personal Computer) or the like of a sales person who conducts business negotiations using a Web conferencing system. Further, the client device 120 is a personal computer or the like of a customer who has received business through a Web conferencing system. The client device 130 is a colleague of a sales person who conducts business negotiations using a Web conferencing system, a PC of a boss, or the like. In FIG. 1, for simplification of the description, in the information processing system 1000, only one client device 110, one client device 120, and one client device 130 are shown, but a plurality of each are included in the information system 1000. It may be. Further, the client device is not limited to the PC, and may be a smartphone, a tablet computer, or the like.

(Outline of processing of server device 100)
The server device 100 associates the video of the sales person and the customer in the business negotiation using the Web conferencing system with the character string data obtained by converting the voice of the sales person and the customer in the business negotiation into a character string, and associates the video with the character string data for each business negotiation. It is stored as data. Then, when the server device 100 receives the search request from the client device 110, the client device 130, and the like, the server device 100 returns the negotiation data corresponding to the search request to the client device of the search request source.

2. 2. Hardware Configuration (1) Hardware Configuration of Server Device 100 FIG. 2 is a diagram showing an example of the hardware configuration of the server device 100. The server device 100 includes a control unit 201, a storage unit 202, and a communication unit 203 as a hardware configuration. The control unit 201 is a CPU (Central Processing Unit) or the like, and controls the entire server device 100. The storage unit 202 is an HDD (Hard Disk Drive), a ROM (Read Only Memory), a RAM (Random Access Memory), or the like, and is a program, data or the like used when the control unit 201 executes processing based on the program. Remember. By executing the processing based on the program stored in the storage unit 202, the control unit 201 realizes the functional configuration of the server device 100 of FIG. 4 described later and the processing of the activity diagrams of FIGS. 5 and 6 described later. Will be done. The communication unit 203 is a NIC (Network Interface Card) or the like, connects the server device 100 to the network 150, and controls communication with another device (for example, a client device 130 or the like).

(2) Hardware Configuration of Client Device 110 FIG. 3 is a diagram showing an example of the hardware configuration of the client device 110. The client device 110 includes a control unit 301, a storage unit 302, an imaging unit 303, an input unit 304, an output unit 305, and a communication unit 306 as hardware configurations. The control unit 301 is a CPU or the like and controls the entire client device 110. The storage unit 302 is an HDD, ROM, RAM, or the like, and stores a program, data or the like used when the control unit 301 executes processing based on the program. The function of the client device 110 is realized by the control unit 301 executing the process based on the program stored in the storage unit 302. The image capturing unit 303 is a camera or the like and images a user of the client device 110. The input unit 304 is a mouse, a keyboard, or the like, and inputs user operations to the control unit 301. Further, the input unit 304 is a microphone or the like, and inputs the user's voice to the control unit 301. The output unit 305 is a display, a speaker, or the like, and displays the processing result or the like of the control unit 201 or outputs it by voice. The communication unit 306 is a NIC or the like, connects the client device 110 to the network 150, and controls communication with another device (for example, the client device 120 or the like).

The hardware configuration of the client device 120 and the client device 130 is the same as the hardware configuration of the client device 110. The function of the client device 120 is realized by the control unit of the client device 120 executing the process based on the program stored in the storage unit of the client device 120. Similarly, the function of the client device 130 is realized by the control unit of the client device 130 executing the process based on the program stored in the storage unit of the client device 130.
3. 3. Functional configuration FIG. 4 is a diagram showing an example of the functional configuration of the server device 100. As shown in FIG. 4, the server device 100 includes a voice recognition unit 401, a storage processing unit 402, a registration unit 403, a search unit 404, and an output control unit 405 as functional configurations.

(Voice recognition unit 401)
The voice recognition unit 401 performs voice recognition based on the voice data of the negotiation and converts it into text. When performing voice recognition, the voice recognition unit 401 recognizes a word that is unknown, based on the data of the word conversion dictionary. In addition, the voice recognition unit 401 classifies the speakers in the negotiation (whether the sales person in charge of sales or the customer receiving sales) based on the waveform of the voice in the negotiation, and the character string for each speaker ( Convert to text). Further, the voice recognition unit 401 may analyze the converted character string and classify the speakers in the negotiation based on the analysis result.

(Memory processing unit 402)
The storage processing unit 402 associates the video data of the negotiation with the text data obtained by converting the voice data of the negotiation into text, and stores the negotiation data in a storage area such as the storage unit 202. In addition, information on the date and time when the negotiation was held, information on the person in charge of sales, information on the customer, and the like are added to the negotiation data and stored in the storage unit 202 and the like.

(Registration Department 403)
The registration unit 403 registers a new word in the word conversion dictionary based on the registration request. More specifically, the registration unit 403 receives a registration request from an external terminal, and registers a new word in the word conversion dictionary based on the received registration request. The external terminal is, for example, a client device 110, a client device 130, or the like. For example, a sales person who has conducted a business negotiation using a Web conferencing system via a client device 110 registers a company name, a service name, and the like in order to register the company name and the service name that appear during the business. The request is transmitted to the server device 100.

(Search unit 404)
The search unit 404 searches for negotiation data based on the search conditions corresponding to the search request, and acquires the negotiation data corresponding to the search conditions as a search result. The search request is transmitted from an external device (for example, client device 110, client device 130, etc.). The search request includes, for example, date and time information of the negotiation to be searched, information of the person in charge of sales, keywords included in the negotiation data, and the like. Further, when a predetermined button is selected on the screen displayed on the output unit 305 of the external device, the search unit 404 searches for the negotiation data including the set word. Here, the predetermined button is, for example, a "word that leads to the next business negotiation" button. In addition, the set words are, for example, words that lead to the next business negotiation. In the search unit 404, when the "words leading to the next negotiation" button is selected on the screen displayed on the output unit 305 of the external device, the search unit 404 includes the words described in the setting file as the words leading to the next negotiation. Search the data. Here, the contents of the setting file may be set by the administrator. Further, the server device 100 may analyze the negotiation data leading to the next negotiation, identify a word contained in a large amount in such negotiation data, and add or update the word to the setting file.

(Output control unit 405)
The output control unit 405 outputs a search result screen including the search result by the search unit 404 to the search request source. More specifically, the output control unit 405 displays a search result screen that summarizes the number of cases in the opportunity data that include words that lead to the next opportunity, together with the identification information that identifies the opportunity data, for each opportunity. Output to the search request source.

4. Information processing FIG. 5 is an activity diagram showing an example of information processing related to registration of the server device 100.
In A501, the voice recognition unit 401 determines whether or not a registration request for data including a moving image of a business negotiation has been received. The server device 100 provides the function of the Web conferencing system. When the business negotiation using the Web conference is completed, the Web conference system transmits a registration request for data including a moving image of the business negotiation to the voice recognition unit 401. When the voice recognition unit 401 determines that the registration request for the data including the video of the negotiation has been received, it recommends the process to S502, and when it determines that the registration request for the data including the video of the negotiation has not been received, the voice recognition unit 401 performs the process of A501. repeat.

In A502, the voice recognition unit 401 performs voice recognition based on the voice data included in the moving image and converts it into text. Here, the voice recognition unit 401 classifies the speakers in the negotiation (whether the sales person who is in business or the customer who is in business) based on the waveform of the voice data, and converts it into a text for each speaker. To do. Further, the voice recognition unit 401 may analyze the converted text and classify the speakers in the negotiation based on the analysis result.

In A503, the storage processing unit 402 associates the video data of the negotiation with the text data obtained by converting the voice data of the negotiation into text and stores it in the storage area such as the storage unit 202 as the negotiation data. In addition, information on the date and time when the negotiation was held, information on the person in charge of sales, information on the customer, and the like are added to the negotiation data and stored in the storage unit 202 and the like.

In A505, the registration unit 403 determines whether or not a registration request has been received from the external terminal. When the registration unit 403 determines that the registration request has been received, it recommends the process to A506, and when it determines that the registration request has not been received, the registration unit 403 repeats the process of A505.

In A506, the registration unit 403 registers the word included in the registration request in the word conversion dictionary. Words included in the registration request include company names and service names that appear in business negotiations using the Web conferencing system.

FIG. 6 is an activity diagram showing an example of information processing related to the search in the server device 100.
In A601, the search unit 404 determines whether or not a search request has been received from the external terminal. If the search unit 404 determines that the search request has been received, the process proceeds to A602, and if it determines that the search request has not been received, the search unit 404 repeats the process of A601.

In A602, the search unit 404 searches the negotiation data based on the search condition corresponding to the search request, and acquires the negotiation data corresponding to the search condition as the search result. Here, if the search request is a search request sent from an external terminal by selecting the "words leading to the next opportunity" button, this search request includes the "words leading to the next opportunity" button. Contains information indicating that was selected. When the search request contains information indicating that the "word leading to the next opportunity" button is selected, the search unit 404 sets the search condition corresponding to the search request as a word leading to the next opportunity. Acquire the word described in and search the opportunity data including the acquired word. Then, the search unit 404 acquires the negotiation data corresponding to the search condition as the search result.

In A603, the output control unit 405 generates a search result screen in which the number of negotiation data including words leading to the next negotiation is summarized for each negotiation together with the identification information for identifying the negotiation data.
In A604, the output control unit 405 controls to output the search result screen generated by A603 to the output unit 305 of the requesting client device. FIG. 7 is a diagram showing an example of the search result screen. As shown in FIG. 7, the search result screen includes the number of cases 701 and 702 that include words leading to the next business negotiation as the search result. In the example of FIG. 7, the company side (sales person in charge side) and the customer side are aggregated and described separately, but the company side alone may be used. Further, on the search result screen, as shown in 703 and 704, the results are summarized in units of business negotiations. Further, "0001" for identifying 703 is described in the upper left of 703, and "0002" for identifying 704 is described in the upper left of 704. In the example of FIG. 7, as the next connected words, "reply by next week", "next time", "next meeting" and the like are shown. In addition, these words may be displayed separately from other words on the screen of FIG. 7. Examples of distinction include changing colors, enlarging fonts, and changing typefaces.

The play button 705 is an example of a moving image object for playing a moving image of a conversation in a part where words leading to the next business negotiation appear in the business negotiation. When the play button 705 is selected, the output control unit 405 controls the output unit 305 of the client device to play a moving image of the conversation in which the words leading to the next business negotiation appear in the business negotiation.
Further, the output control unit 405 controls the conversations in the 703 and 704 so that the conversation between the sales person and the customer can be identifiable. For example, the output control unit 405 may change the color of the conversation and display it in an identifiable manner, or may change the color of the play button and display it in an identifiable manner.

5. Effect of the first embodiment According to the first embodiment, since a new word is registered in the word conversion dictionary based on the registration request, it is possible to provide a technique for improving the accuracy of voice recognition. Further, it is shown in FIG. The search results are displayed together for each opportunity. Therefore, for example, even if you want to search for a word that leads to the next business negotiation during a business negotiation, it is easy to understand who is saying what kind of next business word to which customer. In addition, it is easy to understand how many times a sales person says such a word in one business negotiation in the content where the speakers are separated. In addition, the words can be grasped in text, and the video can be immediately played back to confirm what kind of context or scene the word was said in.

<Modification example>
A modified example will be described. Based on the search results, the output control unit 405 generates a list screen that sequentially displays the sales person who said the next most connected word in the negotiation unit, and displays it in the output unit 305 of the requesting client device. You may control it.
FIG. 8 is a diagram showing an example of a list screen. As shown in FIG. 8, on the list screen, information on the sales person who said the next most connected word in each opportunity is displayed in order. When the confirmation button 801 is selected, the output control unit 405 controls the output unit 305 of the requesting client device to display the details of the selected information. For example, the output control unit 405 controls to display the information in the upper half of FIG.

According to this modification, it is possible to display a list of sales representatives who used the next word in the negotiation in order of the number of times. Also, from the list, you can display a screen that can play the actual video of the sales person who uses the word most. Then, by playing back the video from the screen and checking it, it can be used for education of other sales staff.

(Additional note)
The invention may be provided in each of the following aspects.
In the information processing device, the registration unit receives the registration request from an external terminal and registers a new word in the word conversion dictionary based on the received registration request.
The information processing device has a storage processing unit, and the storage processing unit stores the moving image data of the negotiation and the text data obtained by converting the voice data of the negotiation into text in a storage area as the negotiation data in association with each other. , Information processing equipment.
The information processing device has a search unit and an output control unit, and the search unit searches for the negotiation data based on the search conditions corresponding to the search request, and searches for the negotiation data corresponding to the search conditions. An information processing device that acquires as a search result and outputs a search result screen including the search result by the search unit to a search request source.
In the information processing device, the search unit is an information processing device that searches for business negotiation data including a set word.
In the information processing device, the word is an information processing device that leads to the next business negotiation.
In the information processing device, the output control unit summarizes the number of cases in which the word related to the next business negotiation is included in the business negotiation data together with the identification information for identifying the business negotiation data for each business negotiation result. An information processing device that outputs a screen to the search request source.
An information processing method executed by an information processing apparatus, which includes a first step and a second step. In the first step, voice recognition is performed based on the voice data of a business negotiation, and the text is converted into text. When performing voice recognition, if there is an unrecognizable word, the word is recognized based on the data of the word conversion dictionary, and in the second step, a new word is registered in the word conversion dictionary based on the registration request. Information processing method.
A program for making a computer function as each part of the information processing apparatus.
Of course, this is not the case.

For example, it may be provided as a computer-readable non-temporary storage medium for storing the above-mentioned program.
Further, the above-described embodiments and modifications may be arbitrarily combined and carried out.

Finally, various embodiments of the present invention have been described, but these are presented as examples and are not intended to limit the scope of the invention. The novel embodiment can be implemented in various other embodiments, and various omissions, replacements, and changes can be made without departing from the gist of the invention. The embodiments and modifications thereof are included in the scope and gist of the invention, and are included in the scope of the invention described in the claims and the equivalent scope thereof.

100: Server device 110: Client device 120: Client device 130: Client device 150: Network 201: Control unit 202: Storage unit 203: Communication unit 301: Control unit 302: Storage unit 303: Imaging unit 304: Input unit 305: Output Unit 306: Communication unit 401: Voice recognition unit 402: Memory processing unit 403: Registration unit 404: Search unit 405: Output control unit 701: Number of cases 705: Play button 801: Confirmation button 1000: Information processing system

Claims (5)

It is an information processing device
It has a voice recognition unit, a registration unit, a storage processing unit, a search unit, and an output control unit .
The voice recognition unit
Performs voice recognition based on the voice data of business negotiations, converts it into text, and
When performing the voice recognition, if there is a word that cannot be recognized, the word is recognized based on the data of the word conversion dictionary.
The registration unit based on the registration request, and registers a new word to the word conversion dictionary,
The storage processing unit associates the video data of the negotiation with the text data obtained by converting the voice data of the negotiation into text and stores it in the storage area as the negotiation data.
The search unit searches the negotiation data based on the search conditions corresponding to the search request, acquires the negotiation data corresponding to the search conditions as the search result, and obtains the negotiation data.
The output control unit outputs a search result screen including the search result by the search unit to the search request source.
The search unit searches for opportunity data that includes words that lead to the next opportunity.
Information processing device. In the information processing apparatus according to claim 1,
The registration unit receives the registration request from an external terminal, and registers a new word in the word conversion dictionary based on the received registration request.
Information processing device. In the information processing apparatus according to claim 1 or 2 .
The output control unit displays the search result screen, which summarizes the number of cases in which the word related to the next business negotiation is included in the business negotiation data, together with the identification information for identifying the business negotiation data, for each business negotiation from the search request source. Output to
Information processing device. It is an information processing method executed by an information processing device.
Includes a first step, a second step, a third step, a fourth step, and a fifth step .
In the first step, voice recognition is performed based on the voice data of the negotiation, the text is converted, and if there is a word that cannot be recognized when the voice recognition is performed, the word is recognized based on the data of the word conversion dictionary.
In the second step, a new word is registered in the word conversion dictionary based on the registration request.
In the third step, the moving image data of the negotiation and the text data obtained by converting the voice data of the negotiation into text are associated and stored in the storage area as the negotiation data.
In the fourth step, the negotiation data is searched based on the search conditions corresponding to the search request, and the negotiation data corresponding to the search conditions is acquired as the search result.
In the fifth step, the search result screen including the search result is output to the search request source.
In the fourth step, the negotiation data including the words leading to the next negotiation is searched.
Information processing method. A program for causing a computer to function as each part of the information processing apparatus according to any one of claims 1 to 3 .