JP2004295805A - Keyword extraction device and automatic application starting device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a keyword extraction device for extracting an e-mail address and a uniform resource locator (URL) as character string data (a keyword) from voice data, and also to provide an automatic application starting device provided with a computer for starting an application program corresponding to the sort of the keyword. <P>SOLUTION: This keyword extraction device comprises: a character string extraction means for extracting a specific character string through voice recognition of voice data; a voice recognition means for converting a voice of prescribed length including the specific character string extracted from the voice data, into a character string; and a character string correcting means for correcting the character string to generate the keyword. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a keyword extracting device for extracting an e-mail address and a URL (Uniform Resource Locator) from audio data as character string data (keywords), and an automatic application starter including a computer for starting an application program corresponding to the type of the keyword. Equipment related.
[0002]
[Prior art]
[Patent Document 1] Japanese Patent Application Laid-Open No. 10-164529 With the spread of services on the Internet such as e-mail and World Wide Web (WWW), e-mail addresses and HTTP (HyperText Transfer Protocol) URLs are being broadcast during broadcasting on televisions and radios. Is being talked about. The viewer can send an e-mail to the e-mail address spoken during the broadcast or access the URL spoken during the broadcast using a web browser.
[0003]
However, in this case, it is not easy to send an e-mail to the e-mail address or access the URL unless the broadcast is recorded or recorded, or the e-mail address or URL spoken during the broadcast is recorded in a memo. Did not. In particular, if the viewer hears an e-mail address or URL while driving, it is not easy to send an e-mail to this e-mail address or access the URL because notes cannot be taken on the spot. Was.
[0004]
As a method for solving such a problem, there has been proposed a method of including an electronic mail address and a URL in digital broadcast reception data, such as a television broadcasting method disclosed in Patent Document 1. However, the method disclosed in Patent Literature 1 needs to be a digital method that enables broadcasting to transmit an e-mail address or URL as character string data.
[0005]
[Problems to be solved by the invention]
In order to solve the above problem, the present invention includes a keyword extracting device for extracting an e-mail address and a URL from audio data as character string data (keyword), and a computer for starting an application program corresponding to the type of the keyword. It is an object to provide an automatic application starting device.
[0006]
[Means for Solving the Problems]
In order to achieve the above object, a keyword extracting apparatus according to the present invention includes a character string extracting unit that performs voice recognition on voice data to extract a specific character string, and a predetermined length including a specific character string extracted from the voice data. And a character string correcting means for converting the character string into a character string and generating a keyword by correcting the character string.
[0007]
According to the present invention, a specific character string such as "@" (at mark) which is always used for an e-mail address and "www"("double") which is frequently used for an HTTP URL is used as audio data. , A voice of a predetermined length including a specific character string is extracted as a character string, and keywords such as an e-mail address and a URL can be extracted from the character string. That is, according to the present invention, it is possible to automatically obtain a keyword as character string data from an audio output from a radio or a television.
[0008]
Further, according to the present invention, there is provided an automatic application activation device including the keyword extraction device and a computer connected to the keyword extraction device, wherein the computer is configured to perform a function according to a type of the keyword generated by the keyword extraction device. A program for starting the application program is provided.
[0009]
Therefore, according to the automatic application starting device of the present invention, when a keyword is extracted by the keyword extracting device, an application (e-mail client, web browser, or the like) corresponding to the type of the keyword is automatically started, and the viewer is minimally viewed. With this operation, an e-mail can be sent and a URL can be accessed.
[0010]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 is a block diagram of an automatic application starting device according to an embodiment of the present invention. The automatic application activation device 1 includes a broadcast receiver 100, a keyword extraction device 200, a computer 300, an audio cable C1, and a data cable C2.
[0011]
The broadcast receiver 100 has a receiving unit 101, an antenna 102, and a line output terminal 111. The audio signal received by the antenna 102 is sent to the receiving unit 101. The receiving section demodulates the audio signal to generate audio data. The generated audio data is output from the line output terminal 111.
[0012]
The keyword extracting device 200 includes an A / D converter 201, a voice storing unit 202, a character string extracting unit 203, a keyword extracting unit 204, a keyword storing unit 205, a keyword editing unit 206, a keyword transmitting unit 207, It has a controller 208, a line input terminal 211, and a data output terminal 212.
[0013]
The line output terminal 111 of the broadcast receiver 100 and the line input terminal 211 of the keyword extracting device 200 are connected by an audio cable C1. The audio data generated by the receiving unit 101 is sent to the A / D converter 201 via the line input terminal 211.
[0014]
The A / D converter 201 quantizes the audio data and sends it to the audio storage unit 202.
[0015]
The audio storage unit 202 generates an audio data file in which the quantized audio data is collected for each fixed size, and sequentially stores the audio data files in the audio data storage memory 202a in the audio storage unit 202. After the generation, the audio data file for which a predetermined time has elapsed is deleted from the audio data storage memory 202a.
[0016]
The voice data file stored in the voice data storage memory 202a is read by the character string extraction unit 203. The character string extraction unit 203 generates a specific character string such as "@" (at mark) which is always included in the e-mail address in the voice data file, or "www"("double") included in the HTTP URL. It is determined whether or not a data sequence corresponding to the waveform is included, and the determination result is transmitted to the keyword extracting unit 204.
[0017]
The keyword extraction unit 204 recognizes the voice data file stored in the voice data storage memory 202a by using the determination result of the character string extraction unit 203, and may include an e-mail address or an HTTP URL. To extract character string data with high The character string data generated by the keyword extraction unit 204 is stored in the keyword storage unit 205. The extraction of the character string data described above is performed by the controller 208 controlling the character string extracting unit 203 and the keyword extracting unit 204. The specific operation of the controller 208 will be described later.
[0018]
The character string data generated by the keyword extracting unit 204 may include character strings other than email addresses and HTTP URLs. In addition, an erroneous detection of voice recognition may cause an e-mail address or an HTTP URL (keyword) to be recognized as an incorrect character string. The keyword editing unit 206 is used to extract only keywords from the character string data and further correct the keywords. The keyword editing unit 206 includes a monitor 206a for displaying character string data stored in the keyword storage unit 205, and character string input means 206b such as a keyboard. The user of the automatic application activation device 1 can correct the character string data stored in the keyword storage unit 205 by operating the character string input unit 206b while checking the character string data displayed on the monitor 206a.
[0019]
The character string data corrected by the keyword editing unit 206, that is, the keyword, is read by the keyword transmitting unit 207. This keyword is output from the data output terminal 212.
[0020]
The computer 300 has a CPU 301, a keyword receiving unit 302, a memory 303, a storage 304, a monitor 306, a keyboard 308, and a data input terminal 311.
[0021]
The data output terminal 212 of the keyword extracting device 200 and the data input terminal 311 of the computer 300 are connected by a data cable C2. The keyword is sent to the keyword receiving unit 302 via the data input terminal 311.
[0022]
When a keyword is input to the keyword receiving unit, the CPU 301 determines whether the keyword is an email address or an HTTP URL. That is, when the keyword includes “@” unique to the email address, the CPU 301 determines that the keyword is the email address. On the other hand, if the keyword includes “$”, the CPU 301 determines that the keyword is an HTTP URL.
[0023]
If the CPU 301 determines that the keyword is an e-mail address, it activates an e-mail client. At this time, the e-mail client displays an e-mail creation screen, and a keyword is input to the destination of the e-mail (To :). Therefore, the user of the automatic application starting device 1 does not need to manually input the e-mail address.
[0024]
When the CPU 301 determines that the keyword is an HTTP URL, it activates a web browser. At this time, a keyword is entered in the URL entry field of the web browser. Therefore, the user of the automatic application starting device 1 does not need to manually input the URL.
[0025]
Note that application programs such as an e-mail client and a web browser are stored in the storage 304. The CPU 301 loads the application from the storage 304 to the memory 303 and starts the application. The execution result of the application is displayed on the monitor 306, and the user of the automatic application activation device 1 operates the application using the keyboard 308.
[0026]
As described above, according to the present embodiment, when an e-mail address or an HTTP URL is pronounced during broadcasting, a character string including the e-mail address or the HTTP URL is automatically recorded. Therefore, the user of the automatic application activation device 1 can obtain the e-mail address and the HTTP URL by correcting the character string. Furthermore, according to the present embodiment, an application (e-mail client or web browser) corresponding to the obtained e-mail address or HTTP URL is automatically started.
[0027]
In the present embodiment, the broadcast receiver 100 and the keyword extracting device 200 are provided as separate devices. However, the broadcast receiver 100 may have a configuration in which the keyword extracting device 200 is incorporated.
[0028]
The keyword extraction procedure by the keyword extraction device 200 will be described below. FIG. 2 is a flowchart showing a keyword extraction procedure by the character string extraction unit 203 and the keyword extraction unit 204 of the keyword extraction device 200. This flowchart is a program executed by the controller 208.
[0029]
When this flow is executed, step S101 is first executed. In step S101, the controller 208 controls the character string extraction unit 203 to detect whether or not a certain audio data file includes a data string of a specific waveform. In the present embodiment, the data string of the specific waveform is a specific character string such as "@" (at mark) which is always included in the e-mail address and "www"("double") included in the URL of HTTP. Is a data string corresponding to a waveform when a sound is generated. Next, the process proceeds to step S102.
[0030]
In step S102, the controller 208 determines whether or not the audio data file includes a data string of a waveform when a specific character string is generated. If the audio data file does not include a data string of a waveform when a specific character string is pronounced (S102: NO), the process proceeds to step S101, and the operation of step S101 is performed on another audio data file. On the other hand, if it is determined in step S102 that the audio data file contains a data string of a waveform when a specific character string is pronounced (S102: YES), the process proceeds to step S103.
[0031]
In step S103, the character string extracting unit 203 determines “the file name of the audio data file including the data string of the specific waveform” and “the start address where the data string of the specific waveform starts in the audio data file” ( Hereinafter, these are collectively referred to as keyword information).
[0032]
When the keyword information is transmitted from the character string extracting unit 203, the keyword extracting unit 204 extracts the audio data file stored in the audio data storing memory 202a from the audio data file for a specific time before and after the head address at which the specific waveform data sequence starts. Extract the audio data of Further, the character string extracting unit 203 generates character string data by performing voice recognition on the extracted voice data. In a few seconds including when the specific character string is pronounced, it is highly possible that an e-mail address or an HTTP URL is included. In the present embodiment, the length within a specific time is set so that an e-mail address of a general length or an HTTP URL can be completely contained within a specific time before and after a specific character string is pronounced. (For example, 30 seconds). Next, the process proceeds to step S104.
[0033]
In step S104, the character string data generated by the keyword extraction unit 204 is stored in the keyword storage unit 205. Next, returning to step S101, the operation of step S101 is performed on another audio data file.
[0034]
【The invention's effect】
As described above, according to the present invention, a keyword extracting device that extracts an e-mail address and a URL (Uniform Resource Locator) from audio data as character string data (keyword) and an application program corresponding to the type of keyword are started. An automatic application starting device including a computer is realized.
[Brief description of the drawings]
FIG. 1 is a block diagram of an automatic application starting device according to an embodiment of the present invention.
FIG. 2 is a flowchart of a keyword extracting operation according to the embodiment of the present invention.
[Explanation of symbols]
1 Automatic application activation device 100 Broadcast receiver 200 Keyword extraction device 202 Voice storage unit 203 Character string extraction unit 204 Keyword extraction unit 205 Keyword storage unit 206 Keyword editing unit 207 Keyword transmission unit 208 Controller 300 Computer 301 CPU
302 Keyword receiving unit

Claims (12)

Character string extracting means for recognizing voice data and extracting a specific character string;
Voice recognition means for converting a voice of a predetermined length including the specific character string extracted from the voice data to a character string,
Character string correction means for correcting the character string to generate a keyword;
A keyword extraction device, comprising: The keyword extracting device according to claim 1, wherein the keyword extracting device has a voice input terminal to which the voice data is input. The keyword extraction device is connected to a broadcast receiver via the audio input terminal,
3. The keyword extracting device according to claim 2, wherein the character string extracting unit extracts a specific character string from audio data demodulated by the broadcast receiver. The keyword extracting device is built in a broadcast receiver,
2. The keyword extracting device according to claim 1, wherein the character string extracting unit extracts a specific character string from audio data demodulated by the broadcast receiver. 5. The keyword extracting device according to claim 1, wherein the keyword is an electronic mail address. The keyword extracting device according to claim 5, wherein the specific character string is "$". The keyword extracting device according to claim 1, wherein the keyword is an HTTP URL. The keyword extracting device according to claim 7, wherein the specific character string is "WWW". 9. The keyword extraction device according to claim 1, wherein the keyword extraction device includes a keyword storage unit that stores the keyword generated by the character string correction unit. An automatic application starting device, comprising: the keyword extracting device according to claim 1; and a computer connected to the keyword extracting device.
The automatic application starting device, wherein the computer starts an application program corresponding to a type of a keyword generated by the keyword extracting device. The apparatus of claim 10, wherein the computer activates a mail client when the keyword is an e-mail address. The automatic application activation device according to claim 10, wherein, when the keyword is an HTTP URL, the computer activates a web browser.

Images