JP4257563B2 - Information processing apparatus and method, recording medium, and database construction apparatus - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an information processing apparatus and method. , recoding media And database construction device For example, an information processing apparatus and method suitable for use in constructing a database of information related to music used in television broadcasting, etc. , recoding media And database construction device About.
[0002]
[Prior art]
For example, when a viewer is watching a television broadcast, he / she may want to know attribute information (song name, singer name, etc.) about the music used in the program. Accordingly, there are service organizations that provide attribute information about music used in broadcast programs.
[0003]
[Problems to be solved by the invention]
In such service organizations, a database is constructed by organizing attribute information about music used in television broadcasting in a time series, but most of the process of constructing the database relies on human power. Yes. That is, the database detects a song while a person who is familiar with various songs watches a television broadcast program or the like in real time or watches a recorded program, and detects the song name and singer name of the detected song. It is constructed by accumulating the determined information.
[0004]
As described above, the conventional method of creating a database of attribute information about broadcasted music relies on human power, so it takes a delay of about one day from the time the music is broadcast until it is registered in the database. There was a problem.
[0005]
In addition, the method of recording a broadcasted program and reproducing it to construct a database also records parts other than the music included in the broadcasted program, so the capacity of the recording medium used for recording is effectively utilized. There are not challenges.
[0006]
Furthermore, since it relies on human power, there is a limit to processing capacity, and when the number of broadcast programs further increases, there may be a situation where human resources are insufficient.
[0007]
The present invention has been made in view of such a situation, and an object thereof is to automatically construct a database of attribute information relating to music used in television broadcasting or the like.
[0008]
[Means for Solving the Problems]
The information processing apparatus according to the present invention includes an input unit that inputs a data stream, a supply unit that supplies an input time of the data stream, a storage unit that stores the input time in association with the data stream, and a primary search process. If it is determined whether the input data stream can be processed and if it is determined that it can be processed, A data stream is divided every predetermined period to generate a divided data stream, the type of the divided data stream is determined, and if it is determined to be a predetermined type of data, the feature information of the divided data stream is generated and generated Based on the feature information, the attribute information corresponding to the divided data stream is searched, and when the attribute information corresponding to the divided data stream is obtained as a search result, the searched attribute information and the input time are associated with the divided data stream. , A primary search processing means for performing processing for constructing a search result database, and secondary search pre-processing, In accumulation means The accumulated data stream is read every predetermined period to generate a divided data stream, the type of the divided data stream is determined, and when it is determined as the predetermined type of data, the feature information of the divided data stream is generated, The divided data stream is a divided data stream that is recorded in the search history database by associating the generated feature information and the input time with the divided data stream, and is recorded in the search history database. Secondary search preprocessing means for performing processing for identifying a stream, associating feature information and input time with the specified divided data stream and registering them in the secondary search wait database, and waiting for secondary search as secondary search processing Based on the feature information of the divided data stream registered in the database, the divided data When attribute information corresponding to a stream is searched and attribute information corresponding to a divided data stream is obtained as a search result, the searched attribute information and input time are associated with the divided data stream and additionally registered in the search result database. Secondary search processing means for performing the processing.
[0009]
An information processing method according to the present invention includes an input step of inputting a data stream by an input unit, a supply step of supplying an input time of the data stream by a supply unit, and an input time associated with the data stream by an accumulation unit. By the accumulation step to accumulate and the primary search processing means, If it is determined whether the input data stream can be processed and if it is determined that it can be processed, A data stream is divided every predetermined period to generate a divided data stream, the type of the divided data stream is determined, and if it is determined to be a predetermined type of data, the feature information of the divided data stream is generated and generated Based on the feature information, the attribute information corresponding to the divided data stream is searched, and when the attribute information corresponding to the divided data stream is obtained as a search result, the searched attribute information and the input time are associated with the divided data stream. , By a primary search processing step for performing processing for constructing a search result database, and a secondary search preprocessing means, In accumulation means The accumulated data stream is read every predetermined period to generate a divided data stream, the type of the divided data stream is determined, and when it is determined as the predetermined type of data, the feature information of the divided data stream is generated, The divided data stream is a divided data stream that is recorded in the search history database by associating the generated feature information and the input time with the divided data stream, and is recorded in the search history database. A secondary search preprocessing step for performing a process of specifying a stream, associating feature information and an input time with the specified divided data stream and registering it in the secondary search waiting database, and a secondary search by the secondary search processing means Based on the feature information of the divided data stream registered in the waiting database When attribute information corresponding to a data stream is searched and attribute information corresponding to a divided data stream is obtained as a search result, the searched attribute information and input time are associated with the divided data stream and added to the search result database. And a secondary search processing step for performing a registration process.
[0010]
The program of the present invention includes an input step for inputting a data stream, a supply step for supplying the input time of the data stream, an accumulation step for storing the data stream in association with the input time, and a primary search process. If it is determined whether the input data stream can be processed and if it is determined that it can be processed, A data stream is divided every predetermined period to generate a divided data stream, the type of the divided data stream is determined, and if it is determined to be a predetermined type of data, the feature information of the divided data stream is generated and generated Based on the feature information, the attribute information corresponding to the divided data stream is searched, and when the attribute information corresponding to the divided data stream is obtained as a search result, the searched attribute information and the input time are associated with the divided data stream. As a primary search processing step for performing a process for constructing a search result database and a secondary search pre-processing, In accumulation means The accumulated data stream is read every predetermined period to generate a divided data stream, the type of the divided data stream is determined, and when it is determined as the predetermined type of data, the feature information of the divided data stream is generated, The divided data stream is a divided data stream that is recorded in the search history database by associating the generated feature information and the input time with the divided data stream, and is recorded in the search history database. A secondary search pre-processing step for performing processing for specifying a stream, associating feature information and input time with the specified divided data stream and registering it in the secondary search waiting database, and waiting for secondary search as the secondary search processing Based on the feature information of the divided data stream registered in the database, the divided data When attribute information corresponding to a divided data stream is obtained as a search result, the attribute information corresponding to the divided data stream is associated with the divided data stream and added to the search result database. Causing the computer of the information processing apparatus to execute processing including a secondary search processing step for performing processing of registration , Computer readable The program is recorded.
[0011]
The database construction apparatus according to the present invention includes a reading means for reading content data recorded on a package medium, an extracting means for extracting feature information of the read content data, and feature information extracted as attribute information of the content data. A construction means for constructing a content database in association with each other, an input means for inputting a data stream, a supply means for supplying an input time of the data stream, a storage means for associating an input time with the data stream, and 1 As the next search process, If it is determined whether the input data stream can be processed and if it is determined that it can be processed, A data stream is divided every predetermined period to generate a divided data stream, the type of the divided data stream is determined, and if it is determined to be a predetermined type of data, the feature information of the divided data stream is generated and generated Based on the feature information, the attribute information corresponding to the divided data stream is searched by referring to the content database, and the attribute information corresponding to the divided data stream is obtained as a search result. And a primary search processing means for performing a process of constructing a search result database by associating the input time with each other, In accumulation means The accumulated data stream is read every predetermined period to generate a divided data stream, the type of the divided data stream is determined, and when it is determined as the predetermined type of data, the feature information of the divided data stream is generated, The divided data stream is a divided data stream that is recorded in the search history database by associating the generated feature information and the input time with the divided data stream, and is recorded in the search history database. Secondary search preprocessing means for performing processing for identifying a stream, associating feature information and input time with the specified divided data stream and registering them in the secondary search wait database, and waiting for secondary search as secondary search processing Content based on the feature information of the divided data stream registered in the database Referring to the database, the attribute information corresponding to the divided data stream is searched, and when the attribute information corresponding to the divided data stream is obtained as a search result, the searched attribute information and the input time are associated with the divided data stream. Secondary search processing means for performing a process of additionally registering in the search result database.
[0012]
In the present invention, as the primary search process, When it is determined whether the input data stream is processable and it is determined that it can be processed, A data stream is divided every predetermined period to generate a divided data stream, the type of the divided data stream is determined, and if it is determined to be a predetermined type of data, the feature information of the divided data stream is generated and generated Based on the feature information, the attribute information corresponding to the divided data stream is searched, and when the attribute information corresponding to the divided data stream is obtained as a search result, the searched attribute information and the input time are associated with the divided data stream. Then, process to build the search result database. As secondary search pre-processing, In accumulation means The accumulated data stream is read every predetermined period to generate a divided data stream, the type of the divided data stream is determined, and when it is determined as the predetermined type of data, the feature information of the divided data stream is generated, The divided data stream is a divided data stream that is recorded in the search history database by associating the generated feature information and the input time with the divided data stream, and is recorded in the search history database. A process of specifying a stream and associating the specified divided data stream with the feature information and the input time in the secondary search waiting database is performed. Further, as secondary search processing, attribute information corresponding to the divided data stream is searched based on the characteristic information of the divided data stream registered in the secondary search waiting database, and attribute information corresponding to the divided data stream is obtained as a search result. Is obtained, the attribute information searched and the input time are associated with the divided data stream and additionally registered in the search result database.
[0015]
DETAILED DESCRIPTION OF THE INVENTION
A configuration example of a database construction system to which the present invention is applied will be described with reference to FIG. This database construction system reads audio signals of music recorded on package media such as music CDs (Compact Discs), detects the characteristics of the music, and constructs a music database in association with attribute information such as music titles. The music database (DB) construction device 1 to be detected and the music titles of the music are detected by detecting the music from the program such as a television broadcast and detecting the characteristics of the music. And the like, and a playlist database (DB) construction device 8 that constructs a playlist database in association with the broadcast time.
[0016]
A configuration example of the music database construction device 1 will be described. The input unit 2 receives the audio signal of the music input from the player 7, converts it into a signal in the WAVE format or the like that can be processed by the subsequent type determination unit 3 and outputs it to the type determination unit 3. The type determination unit 3 analyzes an audio signal in the WAVE format, etc., determines a musical feature that can identify the music piece, and outputs it to the conversion unit 4. The conversion unit 4 converts the musical feature determined by the type determination unit 3 into a numerical value, generates feature information of the music, and outputs it to the recording unit 5.
[0017]
The recording unit 5 includes, in the feature information generated by the conversion unit 4, attribute information (music ID, music name, singer name (player name), songwriter name, composer name, distributor of the music CD, The music database is constructed by associating the copyright owner name and the like, and recorded on the recording medium 6.
[0018]
Of the music attribute information, information recorded on a package medium such as a music CD is read and used, and other information is input by the user of the music database construction device 1.
[0019]
The player 7 reads the audio signal of the music recorded on the music CD and outputs it to the input unit 2 of the music database construction apparatus 1. If the attribute information is recorded on the music CD or the like, the player 7 also reads out the attribute information and outputs it to the input unit 2 of the music database construction device 1.
[0020]
Next, a configuration example of the playlist database construction device 8 will be described. The input unit 9 accepts an audio signal such as a television broadcast input from the tuner 18 or the recording / reproducing unit 11 and divides the signal every predetermined period (for example, one minute), and can be processed in the subsequent stage of the type determination unit 12 The signal is converted into a signal in the WAVE format or the like, and is output to the type determination unit 12 together with information on the time (corresponding to the broadcast time of the audio signal) supplied from the time supply unit 10. The input unit 9 also outputs the audio signal from the tuner 18 in association with the time information from the time supply unit 10 to the recording / reproducing unit 11.
[0021]
The time information supply unit 10 is connected to an NTP (Network Time Protocol) server or the like to match a built-in clock, and the clock indicates Time information Is output to the input unit 9. The recording / reproducing unit 11 encodes the audio signal input from the input unit 9 using a predetermined method (for example, MPEG2 method), and records it on a recording medium (not shown) in association with time information input at the same time. . Further, the recording / reproducing unit 11 reads an audio signal and time information encoded from the recording medium, decodes them, and supplies them to the input unit 9.
[0022]
The type determination unit 12 classifies the type of the audio signal input from the input unit 9 for each predetermined period into music, conversation, operation sound of an object, and the like, and outputs the audio signal classified into music to the conversion unit 13. . The conversion unit 13 detects and digitizes the musical features of the audio signal classified as music, and outputs it to the collation unit 14 as feature information. The collation unit 14 collates the feature information input from the conversion unit 13 with the music database constructed by the music database construction device 1 and outputs the collation result to the aggregation unit 15. Specifically, when it is determined that there is a song whose feature information matches in the song database, the attribute information is read from the song database and output to the counting unit 15 together with the time information. On the other hand, when it is determined that there is no music with matching feature information, the collation unit 14 outputs the feature information and time information converted by the conversion unit 13 to the counting unit 15 as a collation result. Whether or not there is a song whose feature information matches in the song database is determined based on whether or not the difference between the digitized feature information is equal to or less than a predetermined threshold.
[0023]
The totaling unit 15 totals the collation results input from the collation unit 14, generates a playlist database, and outputs the playlist database to the recording unit 16. Here, a playlist is a collection of attribute information of music used in a predetermined broadcast channel in time series, and its items include a broadcast station name (channel), broadcast date and time, and a music database. Attribute information (music ID, music name, singer name (performer name), songwriter name, composer name, CD release source, copyright owner name, etc.).
[0024]
The totaling unit 15 also totals processing results input from the input unit 9, the conversion unit 13, the recording medium 17, and the like via a route (not shown), generates a predetermined database (detailed later), and generates the recording unit 16. Output to. The recording unit 16 records the playlist database and the like input from the totaling unit 15 on the recording medium 17.
[0025]
The tuner 18 receives a broadcast signal of television broadcast or radio broadcast, and outputs an audio signal of a predetermined channel to the input unit 9 of the playlist database construction device 8.
[0026]
Next, the music database construction processing of the music database construction device 1 will be described with reference to the flowchart of FIG. In step S <b> 1, the input unit 2 acquires the audio signal of the music from the player 7, converts it into a signal in a WAVE format or the like that can be processed by the subsequent type determination unit 3 and outputs it to the type determination unit 3. In step S <b> 2, the type determination unit 3 analyzes an audio signal such as a WAVE format, determines a musical feature that can identify the music piece, and outputs it to the conversion unit 4. The conversion unit 4 converts the musical feature determined by the type determination unit 3 into a numerical value, generates feature information of the music, and outputs it to the recording unit 5.
[0027]
In step S <b> 3, the recording unit 5 constructs a music database by associating the feature information input from the conversion unit 4 with the attribute information of the music input by a predetermined method, and records it in the recording medium 6.
[0028]
Note that this music database construction process is promptly executed for an existing music CD or the like or a newly released music CD. Also, for commercial songs that are not made into music CDs, the sound source is procured as much as possible and the music database construction process is performed.
[0029]
Next, a first operation example of the playlist database construction device 8 will be described with reference to the flowchart of FIG. In step S <b> 11, the tuner 18 receives a broadcast signal of television broadcast or radio broadcast, and outputs an audio signal of a predetermined channel to the input unit 9 of the playlist database construction device 8.
[0030]
The input unit 9 associates the audio signal input from the tuner 18 with the time information from the time supply unit 10 and outputs it to the recording / reproducing unit 11. The recording / reproducing unit 11 encodes the audio signal sequentially input from the input unit 9 by a predetermined method, and records it on the recording medium in association with the time information input at the same time.
[0031]
In step S12, a primary search process is executed. Details of the primary search processing will be described with reference to the flowchart of FIG.
[0032]
In step S <b> 21, the input unit 9 determines whether or not the signal from the upper stage can be processed after the type determination unit 12. If it is determined that processing is possible, the process proceeds to step S22.
[0033]
In step S <b> 22, the input unit 9 divides the audio signal input from the tuner 18 at predetermined intervals, converts the audio signal into a WAVE format signal, etc., and outputs it to the type determination unit 12 together with time information from the time supply unit 10. .
[0034]
In step S <b> 23, the type determination unit 12 classifies the type of the audio signal for each predetermined period input from the input unit 9. In step S24, the type determination unit 12 determines whether or not the type determination result in step S23 is music, and if the type determination result is determined to be music, the process proceeds to step S25.
[0035]
In step S <b> 25, the type determination unit 12 outputs an audio signal such as a WAVE format that is a processing target to the conversion unit 13. The conversion unit 13 detects and quantifies the musical features of the audio signal classified as music input from the type determination unit 12, and outputs it to the collation unit 14 as feature information. In step S <b> 26, the collation unit 14 collates the feature information input from the conversion unit 13 with the music database constructed by the music database construction device 1, and as a collation result, there is a music whose feature information matches the music database. If it is determined, the attribute information and time information (broadcast time) are output to the counting unit 15.
[0036]
In step S <b> 27, the counting unit 15 compares the collation result (music attribute information and broadcast time) input from the collation unit 14. Time Information) is registered, and a primary search result database is generated and output to the recording unit 16. The recording unit 16 records the primary search result database input from the totaling unit 15 in the recording medium 17.
[0037]
If it is determined in step S21 that the type determination unit 12 and subsequent units cannot process the signal from the upper stage, the process returns to step S13 in FIG.
[0038]
If it is determined in step S24 that the type determination result is not music, the process returns to step S21, and the subsequent processing is repeated.
[0039]
Returning to FIG. In step S13, secondary search pre-processing is executed. This secondary search in front The process is executed in parallel with the primary search process in step S12. like May be.
[0040]
Details of the secondary search pre-processing will be described with reference to the flowchart of FIG. In step S31, the recording / reproducing unit 11 records in the recording medium on which the audio signal encoded in step S11 of FIG. 3 is recorded, whether there is an audio signal in a period not yet read and supplied to the input unit 9. Determine whether or not. If it is determined that there is an audio signal in a period not supplied to the input unit 9, the process proceeds to step S32.
[0041]
In step S <b> 32, the recording / reproducing unit 11 reads and decodes the audio signal encoded from the recording medium every predetermined period (for example, one minute), and outputs the decoded signal to the input unit 9. The input unit 9 converts the audio signal from the recording / reproducing unit 11 into a signal in the WAVE format or the like and outputs it to the type determining unit 12 together with the time information from the time supply unit 10.
[0042]
In step S <b> 33, the type determination unit 12 classifies the type of the audio signal for each predetermined period input from the input unit 9. In step S34, the type determination unit 12 determines whether or not the type determination result in step S23 is a music piece. If the type determination result is a music piece, the process proceeds to step S35.
[0043]
In step S <b> 35, the type determination unit 12 outputs an audio signal such as a WAVE format that is a processing target to the conversion unit 13. The conversion unit 13 detects and digitizes the musical features of the audio signal classified into music input from the type determination unit 12, and outputs it to the totaling unit 15 as feature information. In step S <b> 36, the totaling unit 15 constructs a music search history database by associating the time when the audio signal determined to be music is broadcast and the feature information thereof, and outputs the music search history database to the recording unit 16. Therefore, in the music search history database, time information of a period determined to be a music and a feature information corresponding thereto are registered in a series of broadcast audio signals. The recording unit 16 records the music search history database input from the totaling unit 15 in the recording medium 17.
[0044]
If it is not determined in step S34 that the type determination result is music, the processes in steps S35 and S36 are skipped.
[0045]
The processing returns to step S31, and the subsequent processing is repeated until it is determined that there is no audio signal in a period not supplied to the input unit 9. If it is determined in step S31 that there is no audio signal for a period not supplied to the input unit 9, the process proceeds to step S37.
[0046]
In step S37, the counting unit 15 compares the music search history database recorded in step S36 with the primary search result database recorded in step S27 of the primary search process, and records the music search history database. Broadcasting time and feature information of an audio signal that is determined not to have a record corresponding to the primary search result database, that is, a period of music but whose attribute information such as the music name is not known Are extracted, a secondary search waiting database in which they are registered is generated and output to the recording unit 17. In step S <b> 38, the recording unit 17 records the secondary search waiting database input from the totaling unit 15 in the recording medium 17.
[0047]
The process returns to step S14 in FIG. In step S14, the secondary search process is started. Details of the secondary search processing will be described with reference to the flowchart of FIG.
[0048]
In step S41, the tabulation unit 15 records in the secondary search waiting database generated in the secondary search pre-processing without the secondary search process (broadcast time of the audio signal determined to be a song and its characteristic information). ) Is determined. If it is determined that there is a record that has not been subjected to the secondary search process, the process proceeds to step S42.
[0049]
In step S <b> 42, the broadcast time of the audio signal determined to be the music of the predetermined period and the characteristic information thereof are read from the secondary search waiting database and supplied to the matching unit 14. In step S <b> 43, the collation unit 14 collates the supplied feature information with the music database constructed by the music database construction device 1, and outputs the collation result to the aggregation unit 15. As a collation result, when it is determined that there is a song whose feature information matches in the song database (when a song name or the like is found), the song attribute information and broadcast time from the song database are output to the counting unit 15. When there is no music with matching feature information (when the song name or the like is not found), the feature information and the broadcast time are output to the totaling unit 15.
[0050]
In step S <b> 44, the totaling unit 15 determines whether or not the song name of the audio signal read from the secondary search waiting database is found based on the collation result input from the collation unit 14. If it is determined that the song name or the like has been found, the process proceeds to step S45. In step S <b> 45, the counting unit 15 generates a secondary search result database including the music attribute information input from the verification unit 14 as the verification result and the broadcast time, and outputs the database to the recording unit 16. The recording unit 16 records the secondary search result database input from the totaling unit 15 in the recording medium 17.
[0051]
If it is determined in step S44 that the music title or the like is not found, the process proceeds to step S46. In step S <b> 46, the totaling unit 15 generates an unsearched database that is input from the collating unit 14 as the collation result and includes the feature information of the audio signal determined to be a music and its broadcast time, and outputs the database to the recording unit 16. The recording unit 16 records the unsearched database input from the totaling unit 15 in the recording medium 17.
[0052]
Thereafter, the process returns to step S41, and the subsequent processing is repeated until it is determined that there is no record that has not been subjected to the secondary search processing in the secondary search wait database. If it is determined that there is no record that has not been subjected to the secondary search process in the secondary search waiting database, the process returns to step S15 in FIG.
[0053]
In step S15, the counting unit 15 combines the record of the primary search result database generated in step S12 and the record of the secondary search result database generated in step S14 to construct a playlist database. Output. The recording unit 16 records the playlist database input from the totaling unit 15 in the recording medium 17.
[0054]
Note that the unsearched database record (audio signal feature information and broadcast time) generated in step S46 is processed by, for example, human power, and if the music can be identified, the information is added to the playlist database. .
[0055]
As described above, according to the first operation example of the playlist database construction device 8, the primary search processing is performed on the real-time audio signal supplied from the tuner 18 to generate the primary search result database. When the real-time processing reaches the limit, the secondary search processing is performed for the period of the audio signal that is determined as the music in the secondary search pre-processing and the song name is not specified in the primary search processing. Since the secondary search result database is generated and the playlist database is constructed by combining them, the classification determination, the feature extraction, etc. are performed twice for each audio signal of the television broadcast. Become. Therefore, it is possible to prevent the music included in the audio signal such as television broadcast from being missed, and the probability of specifying the music name or the like is improved.
[0056]
Next, a second operation example of the playlist database construction device 8 will be described with reference to the flowchart of FIG. In step S51, the tuner 18 receives a broadcast signal of television broadcast or radio broadcast, and outputs an audio signal of a predetermined channel to the input unit 9 of the playlist database construction device 8.
[0057]
The input unit 9 associates the audio signal input from the tuner 18 with the time information from the time supply unit 10 and outputs it to the recording / reproducing unit 11. The recording / reproducing unit 11 encodes the audio signal sequentially input from the input unit 9 by a predetermined method, and records it on the recording medium in association with the time information input at the same time.
[0058]
In step S52, a primary search process is executed. Details of the primary search process will be described with reference to the flowchart of FIG.
[0059]
In step S <b> 61, the input unit 9 determines whether the signal from the upper stage can be processed after the type determination unit 12. If it is determined that processing is possible, the processing proceeds to step S62.
[0060]
In step S <b> 62, the input unit 9 divides the audio signal input from the tuner 18 at predetermined intervals, converts the audio signal into a WAVE format signal, etc., and outputs the signal to the type determination unit 12 together with time information from the time supply unit 10. .
[0061]
In step S <b> 63, the type determination unit 12 classifies the type of the audio signal for each predetermined period input from the input unit 9. In step S64, the type determination unit 12 determines whether or not the type determination result in step S63 is a music piece. If the type determination result is a music piece, the process proceeds to step S65.
[0062]
In step S <b> 65, the type determination unit 12 outputs an audio signal such as a WAVE format that is a processing target to the conversion unit 13. The conversion unit 13 detects and quantifies the musical features of the audio signal classified as music input from the type determination unit 12, and outputs it to the collation unit 14 as feature information. In step S <b> 66, the collation unit 14 collates the feature information input from the conversion unit 13 with the music database constructed by the music database construction device 1, and outputs the collation result to the aggregation unit 15. As a collation result, when it is determined that there is a song whose feature information matches in the song database (when a song name or the like is found), the song attribute information and broadcast time from the song database are output to the counting unit 15. When there is no music with matching feature information (when the song name or the like is not found), the feature information and the broadcast time are output to the totaling unit 15.
[0063]
In step S <b> 67, the counting unit 15 determines whether or not the music name of the audio signal determined to be a music has been found based on the verification result input from the verification unit 14. If it is determined that the song name or the like has been found, the process proceeds to step S68. In step S <b> 68, the tabulation unit 15 generates a primary search result database using the attribute information and broadcast time of the music input as the collation result from the collation unit 14 and outputs them to the recording unit 16. The recording unit 16 records the primary search result database input from the totaling unit 15 in the recording medium 17.
[0064]
If it is determined in step S67 that the music title or the like is not found, the process proceeds to step S69. In step S <b> 69, the totaling unit 15 generates a secondary search wait database that is input from the verification unit 14 as a verification result and includes the feature information of the audio signal determined to be a music and its broadcast time, and outputs the database to the recording unit 16. To do. The recording unit 16 records the secondary search waiting database input from the totaling unit 15 in the recording medium 17.
[0065]
Thereafter, the process returns to step S61, and the subsequent processing is repeated until it is determined that the signal from the upper stage cannot be processed in the type determination unit 12 and thereafter. Similarly, when it is determined in step S64 that the type determination result is not music, the process returns to step S61.
[0066]
If it is determined in step S61 that the signal from the upper stage cannot be processed by the type determination unit 12 or later, the fact that the primary search processing has not been performed on the audio signal after that time is awaiting the secondary search. After being recorded in the database, the process returns to step S53 in FIG. Here, in the secondary search waiting database, as a result of performing the primary search process, it is determined that the song is a song, and the feature information and broadcast time of the audio signal whose song name or the like is not known, and the type determination The broadcast time of the audio signal that has not been subjected to the primary search process because the part 12 and later cannot be processed is recorded.
[0067]
In step S53, a secondary search process is executed. Details of the secondary search process will be described with reference to the flowchart of FIG.
[0068]
In step S71, the counting unit 15 records in the secondary search waiting database generated by the primary search process a record that has not been subjected to the secondary search process (broadcast time of the audio signal determined to be a song and its characteristic information, 1 It is determined whether or not there is an audio signal broadcast time that has not been subjected to the next search process. If it is determined that there is a record that has not been subjected to the secondary search process, the process proceeds to step S92.
[0069]
In step S <b> 72, the broadcast time of the audio signal determined to be a song and its characteristic information are read from the secondary search waiting database and supplied to the matching unit 14. When the broadcast time of the audio signal that has not been subjected to the primary search process is read from the secondary search waiting database, the audio signal recorded in step S52 is read at predetermined intervals based on the information. The feature information is extracted, converted into feature information, and supplied to the collation unit 14 together with the broadcast time only for a period in which the information is read out, determined by the type determination unit 12 and determined as a music piece there.
[0070]
In step S <b> 73, the collation unit 14 collates the supplied feature information with the music database constructed by the music database construction device 1, and outputs the collation result to the aggregation unit 15. In step S <b> 74, the totaling unit 15 determines whether or not the music name of the audio signal to be processed has been found based on the collation result input from the collation unit 14. If it is determined that the song name or the like has been found, the process proceeds to step S75. In step S <b> 75, the aggregation unit 15 generates a secondary search result database using the music attribute information and the broadcast time input from the verification unit 14 as the verification result, and outputs the database to the recording unit 16. The recording unit 16 records the secondary search result database input from the totaling unit 15 in the recording medium 17.
[0071]
If it is determined in step S74 that the music title or the like is not found, the process proceeds to step S76. In step S <b> 76, the totaling unit 15 generates an unsearched database including the feature information of the audio signal determined as the music and the broadcast time, which are input from the verification unit 14 as the verification result, and outputs the database to the recording unit 16. The recording unit 16 records the unsearched database input from the totaling unit 15 in the recording medium 17.
[0072]
Thereafter, the process returns to step S71, and the subsequent processing is repeated until it is determined that there is no record that has not been subjected to the secondary search processing in the secondary search waiting database. If it is determined that there is no record that has not been subjected to the secondary search process in the secondary search waiting database, the process returns to step S54 in FIG.
[0073]
In step S54, the counting unit 15 combines the record of the primary search result database generated in step S52 and the record of the secondary search result database generated in step S53 to construct a playlist database, and stores it in the recording unit 16. Output. The recording unit 16 records the playlist database input from the totaling unit 15 in the recording medium 17.
[0074]
Note that the unsearched database record (audio signal feature information and broadcast time) generated in step S76 is processed by, for example, human power, and if the music can be specified, the information is added to the playlist database. .
[0075]
As described above, according to the second operation example of the playlist database construction device 8, the type determination, the feature detection, etc. are performed only once for the audio signal supplied from the tuner 18, Compared to the first operation example described above, the time required for a series of processing can be reduced.
[0076]
In the present embodiment, an example of constructing a playlist database for music included in audio signals of television broadcasts and radio broadcasts has been described. However, the present invention is included in video data such as television broadcasts. The present invention can be applied when creating a playlist database of specific images.
[0077]
In addition, the present invention is not limited to television broadcasting and radio broadcasting, and can be applied to constructing a playlist database of predetermined types of information included in all streaming data.
[0078]
Incidentally, the series of processes described above or a part of the processes can be executed by hardware, but can also be executed by software. When a series of processing is executed by software, a program constituting the software may execute various functions by installing a computer incorporated in dedicated hardware or various programs. For example, it is installed from a recording medium in a general-purpose personal computer or the like.
[0079]
FIG. 10 shows a configuration example of a general-purpose personal computer. This personal computer includes a CPU (Central Processing Unit) 21. An input / output interface 25 is connected to the CPU 21 via the bus 24. The input / output interface 25 includes an input unit 26 including an input device such as a keyboard and a mouse, an output unit 27 that outputs processing results to the outside, a communication unit 28 that communicates data via a network such as the Internet, and a hard disk that stores programs. A drive 29 and a drive 30 for reading / writing data from / to the magnetic disk 31 to the semiconductor memory 34 are connected. A ROM (Read Only Memory) 22 and a RAM (Random Access Memory) 23 are connected to the bus 24.
[0080]
The recording medium on which the program is recorded is distributed to provide the program to the user separately from the personal computer, and the magnetic disk 31 (including the floppy disk) on which the program is recorded, the optical disk 32 (CD-ROM ( Package media, such as Compact Disc-Read Only Memory (DVD), DVD (Digital Versatile Disc), magneto-optical disc 33 (including MD (Mini Disc)), or semiconductor memory 34, or programs are temporary or permanent It consists of a ROM, hard disk, etc. The program is stored in the recording medium using a wired or wireless communication medium such as a local area network, the Internet, or digital satellite broadcasting via an interface such as a router or a modem as necessary.
[0081]
A program for causing the personal computer to execute the processing as described above is supplied to the personal computer in a state stored in a recording medium, read by the drive 30, and installed in the hard disk drive 29. The program installed in the hard disk drive 29 is loaded from the hard disk drive 29 to the RAM 23 and executed by a command of the CPU 21 corresponding to a command from the user input to the input unit 26.
[0082]
In the present specification, the step of describing the program recorded in the recording medium is not limited to the processing performed in time series according to the described order, but is not necessarily performed in time series, either in parallel or individually. The process to be executed is also included.
[0083]
Further, in this specification, the system represents the entire apparatus constituted by a plurality of apparatuses.
[0084]
【The invention's effect】
As described above, the present invention In According to De It is possible to automatically construct a database of attribute information relating to a predetermined type of data included in the data stream.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration example of a database construction system to which the present invention is applied.
FIG. 2 is a flowchart for explaining music database construction processing of the music database construction device 1;
FIG. 3 is a flowchart illustrating a first playlist database construction process of the playlist database construction device 8;
FIG. 4 is a flowchart for explaining details of a primary search process in step S12 of FIG.
FIG. 5 is a flowchart illustrating details of secondary search pre-processing in step S13 of FIG. 3;
6 is a flowchart illustrating details of a secondary search process in step S14 of FIG.
FIG. 7 is a flowchart illustrating a second playlist database construction process of the playlist database construction device 8;
FIG. 8 is a flowchart illustrating details of a primary search process in step S52 of FIG.
FIG. 9 is a flowchart illustrating details of the secondary search process in step S53 of FIG.
FIG. 10 is a block diagram illustrating a configuration example of a general-purpose personal computer.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 Music database construction apparatus, 2 Input part, 3 Type determination part, 4 Conversion part, 5 Recording part, 6 Recording medium, 7 Player, 8 Playlist database construction apparatus, 9 Input part, 10 Time information supply part, 11 Recording / reproduction Section, 12 type determination section, 13 conversion section, 14 collation section, 15 counting section, 16 recording section, 17 recording medium

Claims (6)

In an information processing apparatus for constructing a database related to a predetermined type of data included in a data stream,
Input means for inputting the data stream;
Supply means for supplying an input time of the data stream;
Storage means for storing the input time in association with the data stream;
As the primary search process,
It is determined whether or not the input data stream can be processed, and when it is determined that the data stream can be processed, the data stream is divided every predetermined period to generate a divided data stream,
Determining the type of the divided data stream;
When it is determined that the data is the predetermined type, the feature information of the divided data stream is generated,
Based on the generated feature information, search for attribute information corresponding to the divided data stream,
When attribute information corresponding to the divided data stream is obtained as a search result, a primary search is performed for associating the searched attribute information and the input time with the divided data stream to construct a search result database. Processing means;
As secondary search pre-processing,
Read the data stream stored in the storage means every predetermined period to generate a divided data stream,
Determining the type of the divided data stream;
When it is determined that the data is the predetermined type, the feature information of the divided data stream is generated,
Corresponding the generated feature information and the input time to the divided data stream, constructing a search history database,
A divided data stream recorded in the search history database and not recorded in the search result database is specified, and the feature information and the input time are associated with the specified divided data stream Secondary search preprocessing means for performing processing for registration in the secondary search waiting database;
As a secondary search process,
Based on the feature information of the divided data stream registered in the secondary search waiting database, search for attribute information corresponding to the divided data stream;
When attribute information corresponding to the divided data stream is obtained as a search result, a process of additionally registering the searched attribute information and the input time in the divided data stream in association with the input time is performed 2 An information processing apparatus including: next search processing means. The information processing apparatus according to claim 1, wherein the data stream includes at least one of a television broadcast signal and a radio broadcast signal. The information processing apparatus according to claim 1, wherein the predetermined type of data is music data. An input means for inputting a data stream;
Supply means for supplying the input time of the data stream;
Storage means for storing the input time in association with the data stream;
Primary search processing means for performing primary search processing;
Secondary search preprocessing means for performing secondary search preprocessing;
Secondary search processing means for performing secondary search processing,
In the information processing method of the information processing apparatus for constructing a database related to a predetermined type of data included in the data stream,
An input step of inputting the data stream by the input means;
A supplying step of supplying an input time of the data stream by the supplying means;
A storage step of storing the input time in association with the data stream by the storage means;
By the primary search processing means,
It is determined whether or not the input data stream can be processed, and when it is determined that the data stream can be processed, the data stream is divided every predetermined period to generate a divided data stream,
Determining the type of the divided data stream;
When it is determined that the data is the predetermined type, the feature information of the divided data stream is generated,
Based on the generated feature information, search for attribute information corresponding to the divided data stream,
When attribute information corresponding to the divided data stream is obtained as a search result, a primary search is performed for associating the searched attribute information and the input time with the divided data stream to construct a search result database. Processing steps;
By secondary search preprocessing means,
Read the data stream stored in the storage means every predetermined period to generate a divided data stream,
Determining the type of the divided data stream;
When it is determined that the data is the predetermined type, the feature information of the divided data stream is generated,
Corresponding the generated feature information and the input time to the divided data stream, constructing a search history database,
A divided data stream recorded in the search history database and not recorded in the search result database is specified, and the feature information and the input time are associated with the specified divided data stream A secondary search preprocessing step for performing a process of registering in the secondary search waiting database;
By secondary search processing means,
Based on the feature information of the divided data stream registered in the secondary search waiting database, search for attribute information corresponding to the divided data stream;
When attribute information corresponding to the divided data stream is obtained as a search result, a process of additionally registering the searched attribute information and the input time in the divided data stream in association with the input time is performed 2 An information processing method comprising: a next search processing step. A program for controlling an information processing apparatus that builds a database related to a predetermined type of data included in a data stream,
An input step of inputting the data stream;
A supplying step of supplying an input time of the data stream;
An accumulation step of associating and accumulating the input time with the data stream;
As the primary search process,
It is determined whether or not the input data stream can be processed, and when it is determined that the data stream can be processed, the data stream is divided every predetermined period to generate a divided data stream,
Determining the type of the divided data stream;
When it is determined that the data is the predetermined type, the feature information of the divided data stream is generated,
Based on the generated feature information, search for attribute information corresponding to the divided data stream,
When attribute information corresponding to the divided data stream is obtained as a search result, a primary search is performed for associating the searched attribute information and the input time with the divided data stream to construct a search result database. Processing steps;
As secondary search pre-processing,
Read the data stream stored in the storage means every predetermined period to generate a divided data stream,
Determining the type of the divided data stream;
When it is determined that the data is the predetermined type, the feature information of the divided data stream is generated,
Corresponding the generated feature information and the input time to the divided data stream, constructing a search history database,
A divided data stream recorded in the search history database and not recorded in the search result database is specified, and the feature information and the input time are associated with the specified divided data stream A secondary search preprocessing step for performing a process of registering in the secondary search waiting database;
As a secondary search process,
Based on the feature information of the divided data stream registered in the secondary search waiting database, search for attribute information corresponding to the divided data stream;
When attribute information corresponding to the divided data stream is obtained as a search result, a process of additionally registering the searched attribute information and the input time in the divided data stream in association with the input time is performed 2 A recording medium on which a computer-readable program is recorded , which causes a computer of an information processing apparatus to execute processing including a next search processing step. In a database construction device for constructing a database related to a predetermined type of data included in a data stream,
Reading means for reading the content data recorded on the package media;
Extraction means for extracting feature information of the read content data;
Construction means for constructing a content database by associating the extracted feature information with the attribute information of the content data;
An input means for inputting a data stream;
Supply means for supplying an input time of the data stream;
Storage means for storing the input time in association with the data stream;
As the primary search process,
It is determined whether or not the input data stream can be processed, and when it is determined that the data stream can be processed, the data stream is divided every predetermined period to generate a divided data stream,
Determining the type of the divided data stream;
When it is determined that the data is the predetermined type, the feature information of the divided data stream is generated,
Based on the generated feature information, referring to the content database, search for attribute information corresponding to the divided data stream,
When attribute information corresponding to the divided data stream is obtained as a search result, a primary search is performed for associating the searched attribute information and the input time with the divided data stream to construct a search result database. Processing means;
As secondary search pre-processing,
Read the data stream stored in the storage means every predetermined period to generate a divided data stream,
Determining the type of the divided data stream;
When it is determined that the data is the predetermined type, the feature information of the divided data stream is generated,
Corresponding the generated feature information and the input time to the divided data stream, constructing a search history database,
A divided data stream recorded in the search history database and not recorded in the search result database is specified, and the feature information and the input time are associated with the specified divided data stream Secondary search preprocessing means for performing processing for registration in the secondary search waiting database;
As a secondary search process,
Based on the feature information of the divided data stream registered in the secondary search waiting database, the attribute database corresponding to the divided data stream is searched with reference to the content database,
When attribute information corresponding to the divided data stream is obtained as a search result, a process of additionally registering the searched attribute information and the input time in the divided data stream in association with the input time is performed 2 A database construction device comprising: next search processing means.

Images