JP2008242376A - Musical piece introduction sentence generating device, narration adding device, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To generate a liner note on which musical performance contents of an arbitrary musical piece are reflected without troubling a person and to impart a narration. <P>SOLUTION: A musical piece introduction sentence generating device includes: a musical piece introduction sentence database 152 which stores data for generating or specifying one musical piece introduction sentence from characteristic values indicative of a plurality of predetermined kinds of sound characteristics; an input means of inputting musical piece data; a feature specifying means of analyzing the musical piece data received by the input means and specifying characteristic values showing the plurality of kinds of sound characteristics; and an acquisition means of acquiring introduction sentence data showing an introduction sentence for the musical piece represented by the musical piece data received by the input means from the respective characteristic values specified by the feature specifying means and the storage contents of the musical piece introduction sentence database, and outputting the introduction sentence data. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a technique for automatically generating a sentence, and more particularly to a technique for automatically generating a music introduction sentence that is a basis for liner notes and narration.

  For example, for music CDs and records, a booklet with an introductory text called liner notes on which the characteristics of the recorded music are written is enclosed. Also, when broadcasting a song on TV or radio, etc., introducing the characteristics of the song by narration in the prelude section, interlude section, and follower section of the song that do not hinder the appreciation of the song There is. For this reason, some people who enjoy playing music and singing want to create liner notes and narrations about their performance and singing. However, in order to create liner notes and narration that accurately represent the characteristics of the music, a wide range of expertise in music is required, and if you ask a music critic with such expertise to create it. A large amount of money is required. Such a cost burden is difficult for people who play music as a hobby, and it has been difficult for these people to create liner notes etc. for their own music. Therefore, various techniques for solving such problems have been proposed, and the technique disclosed in Patent Document 1 can be cited as an example.

Patent Document 1 discloses a karaoke system that adds and plays back a narration suitable for the type of song at the start of the performance of the requested song. More specifically, this karaoke system includes a music data storage device, a narration data storage device, and a host system that controls the entire karaoke system. The music data storage device stores in advance each piece of music data of a plurality of pieces of music with identification information indicating the performance content of the music, and the narration data storage device associates with each of a plurality of types of performance content. Narration data indicating the narration corresponding to the performance content is stored. Then, the host system reads out the music data of the requested music from the music data storage device and reproduces it, and reads out the narration data corresponding to the performance content of the music from the narration data storage device and reproduces it at the start of the reproduction. .
JP-A-11-167388

  According to the technique disclosed in Patent Document 1, when singing a karaoke song, it becomes possible to add a narration suitable for the karaoke song. For example, for an original song written by a user, No narration can be added. Further, even for existing music that was not created by the user himself, narration is not added if narration data for the music is not prepared in advance. That is, with the technique disclosed in Patent Document 1, it is not possible to add a narration that reflects the performance content of an arbitrary piece of music.

  The present invention has been made in view of the above problems, and provides a technique that enables generation of liner notes and the addition of narration that reflect the performance content of an arbitrary piece of music without requiring human intervention. It is an object.

  In order to solve the above problems, the present invention provides a music introduction sentence database storing data for generating or specifying one music introduction sentence from characteristic values indicating each of a plurality of predetermined acoustic characteristics, , Input means to which music data is input, characteristic specifying means for analyzing the music data received by the input means and specifying characteristic values indicating each of the plurality of types of acoustic characteristics, and specified by the characteristic specifying means Acquisition means for acquiring and outputting introductory text data representing an introductory text for the music represented by the music data received by the input means from each characteristic value and the stored content of the music introductory text database A music introduction sentence generation device characterized by the above.

  In a more preferred aspect, the stored contents of the music introduction sentence database possessed by the music introduction sentence generating device are classified for each genre of music, and the acquisition means stores the music represented by the music data received by the input means. A genre identifier indicating a genre to which the genre belongs, and the genre indicated by the genre identifier input from the outside or the genre of the music specified from each characteristic value specified by the feature specifying means is associated with the music introduction sentence database. Referencing stored data, introductory sentence data about the music represented by the music data received by the input means is obtained.

  In order to solve the above problems, the present invention analyzes an input means for inputting music data and music data received by the input means, and a prelude section, an interlude section or a post of the music represented by the music data. A section specifying means for specifying a performance section, an acquisition means for acquiring introductory text reflecting the characteristics of the music represented by the music data received by the input means, and a text data acquired by the acquiring means A voice synthesizing unit that synthesizes voice data of narration, which is a reading voice of an introduction sentence, from the sentence data, and a voice represented by the voice data synthesized by the voice synthesizing unit is superimposed on the section specified by the specifying unit. A narration adding device comprising: a synthesizing means for synthesizing and outputting the audio data and the music data; Subjected to.

  In a more preferred aspect, the narration adding device includes a song introduction sentence database storing data for generating or specifying one song introduction sentence from characteristic values indicating each of a plurality of predetermined acoustic characteristics, and Analyzing the music data received by the input means, and specifying a characteristic value indicating each of the plurality of types of acoustic characteristics, and the acquiring means is specified by the feature specifying means Introductory text data representing an introductory text about the music represented by the music data is acquired from the characteristic value and the stored contents of the music introductory text database.

  In order to solve the above-mentioned problem, the present invention, the computer device, analyzes the music data input to the computer device, and specifies the characteristic value indicating each of a plurality of predetermined acoustic characteristics Means, contents stored in a music introduction database storing data for generating or specifying one music introduction sentence from characteristic values indicating each of the plurality of types of acoustic characteristics, and each specified by the feature specifying means There is provided a program characterized by functioning as an acquisition means for acquiring and outputting introductory text data representing an introductory text about a music represented by the music data from a characteristic value.

  In order to solve the above-described problem, the present invention analyzes a music data input to the computer device and specifies a prelude section, an interlude section or a postlude section of the music represented by the music data. Section identifying means, acquisition means for acquiring sentence data representing an introductory sentence reflecting the characteristics of the music represented by the music data, and voice of narration that is a reading voice of the introductory sentence represented by the sentence data acquired by the acquiring means Voice synthesis means for synthesizing data from the text data, and the voice data and the music data so that the voice represented by the voice data synthesized by the voice synthesis means is superimposed on the section specified by the specification means. Provided is a program characterized by functioning as a combining means for combining and outputting.

  According to the present invention, there is an effect that it is possible to generate liner notes and give narration that reflects the performance contents of an arbitrary piece of music without manpower.

The best mode for carrying out the present invention will be described below with reference to the drawings.
(A: 1st Embodiment)
(A-1: Configuration)
FIG. 1 is a diagram illustrating a configuration example of a liner notes generating apparatus 10 according to a first embodiment of a music introduction sentence generating apparatus according to the present invention. The liner notes generating device 10 is a computer device connected to a communication network (not shown) such as the Internet, and distributes music data and introductory sentence data representing the liner notes of the music represented by the music data as digital contents. Functions as a content server. In the following description, a case will be described in which time-series sampling data obtained by sampling an audio waveform of a music at a predetermined sampling interval and performing A / D conversion is used as the music data. Of course, data obtained by performing code compression may be used. As shown in FIG. 1, the liner notes generating apparatus 10 includes a control unit 110, a communication interface (hereinafter referred to as “IF”) unit 120, an external device IF unit 130, a volatile storage unit 140, a nonvolatile storage unit 150, and It has a bus 160 that mediates data exchange between these components.

  The control unit 110 is, for example, a CPU (Central Processing Unit). The control unit 110 functions as a control center of the liner notes generating apparatus 10 by operating according to various programs stored in the nonvolatile storage unit 150. The processing executed by the control unit 110 will be clarified later.

  The communication IF unit 120 is, for example, a NIC (Network Interface Card), and is connected to the communication network described above. The communication IF unit 120 receives a communication message (for example, a request message for requesting distribution of digital content: hereinafter, a content distribution request message) transmitted according to a predetermined communication protocol such as HTTP, and the control unit Meanwhile, a communication message delivered from the control unit 110 (for example, a communication message in which digital content and its destination are written: hereinafter, a content distribution message) is transmitted to the destination via the communication network.

  The external device IF unit 130 is, for example, a USB (Universal Serial Bus) interface, and connects a recording medium reading device such as a USB memory or a CD-ROM (Compact Disk Read Only Memory) drive to exchange data. Is to do. The external device IF unit 130 serves as an input unit for inputting music data to the liner notes generating apparatus 10 when new music data is stored in the liner notes generating apparatus 10. Specifically, when a recording medium such as a USB memory in which music data is written is connected to the external device IF unit 130, the control unit 110 reads the music data from the recording medium via the external device IF unit 130. Then, after applying predetermined processing, new digital content is registered by writing in a predetermined area in the nonvolatile storage unit 150. When a recording medium reading device such as a CD-ROM drive is connected to the external device IF unit 130, new music data is read from the recording medium set in the recording medium reading device, and the nonvolatile storage unit 150. What is necessary is just to make the control part 110 perform the process written in.

  The volatile storage unit 140 is, for example, a RAM (Random Access Memory), and is used as a work area when executing various programs. On the other hand, the non-volatile storage unit 150 is a hard disk for storing various data and various programs. As shown in FIG. 1, the nonvolatile storage unit 150 stores various databases such as a music database 151 and a music introduction sentence database 152, and various programs such as an OS program 153, a content distribution program 154, and a content registration program 155. Yes.

  In the music database 151, music data indicating an audio waveform of a music identified by the music identifier in association with a music identifier (for example, character string data representing the name of the music) that uniquely identifies each of the plurality of music. And introductory sentence data which is text data representing a sentence (so-called liner notes) introducing the characteristics of the music. In the present embodiment, these music data and introductory text data are handled as the digital contents described above. In the present embodiment, a case where text data in which character codes of characters constituting the introductory text are arranged in the order of description is used as the introductory text data. For example, mutual conversion with text data such as PDF format is performed. Of course, other data formats that can be used may be used.

  The music introduction sentence database 152 obtains introduction sentence data about music (hereinafter referred to as “music to be processed”) whose audio waveform is represented by music data input via the external device IF unit 130 from the music data. It is a database that is referred to in the execution process of the process to be generated. As shown in FIG. 2, the music introduction sentence database 152 includes a characteristic value indicating the acoustic characteristic of the music specified by analyzing the music data of the music to be processed, and a genre identifier indicating the genre to which the music belongs. The database for uniquely identifying the introductory text of the music, and various modes can be considered as an embodiment thereof, and an example thereof is as follows.

(A) First, introductory sentence data representing an introductory sentence suitable for expressing the characteristics of music having the acoustic characteristics indicated by these characteristic values in association with characteristic values indicating each of the plurality of types of acoustic characteristics. Is stored and an introduction sentence table is provided for each genre of music, and the music introduction sentence database 152 is configured. The above characteristic values include “Audio Energy” indicating the total power (volume) of the sound, “Spectral” indicating the brightness and light / heavy level of the sound.
“Centroid”, “Spectral Tilt” indicating how powerful the sound is, “Spectral Flatness dB” indicating how sharp the sound is, such as the sound of a full sound, “High” indicating the degree of glittering sound
“Frequency Content”, “MFCC” characterizing the outline of the sound spectrum with several numerical values, “Energy Band Ratio” or “Bark Energy Band” that is the ratio of the power of each frequency band to the total power of the sound, The basic "HPCP" indicating harmony, chord feeling, keys, etc. may be used.

  FIG. 3 shows, as the acoustic characteristics, a feature amount (for example, “Audio Energy”) indicating the volume of a song and a feature amount (for example, “HPCP” indicating whether the melody of the song is clean or crisp. It is a figure which shows an example of the storage content (FIG.3 (a)) of the introduction sentence table about a classical music at the time of employ | adopting ") and the storage content (FIG.3 (b)) of the introduction sentence table about pops. In this way, the reason why the above introduction sentence table is provided for each genre of music is that, even if the music characteristics are represented by the same set of characteristic values, the characteristics of the music are expressed if the genre to which the music belongs is different. This is because the preferred introduction style (sentence) and the terms and expressions to be used in the introduction may differ. In FIG. 3, a case has been described in which a common acoustic characteristic (in FIG. 3, volume and melody feeling) is adopted as the acoustic characteristic associated with the introductory sentence data, regardless of the genre of the music. Of course, different acoustic characteristics may be adopted.

(B) As a second embodiment of the music introduction sentence database 152, a keyword table for converting characteristic values indicating each of the plurality of types of acoustic characteristics into keywords such as adjectives is provided for each genre of music, A mode in which the music introduction sentence database 152 is configured by providing, for each genre of music, an introduction sentence template for forming an introduction sentence by embedding these keywords in a predetermined blank portion. The reason why a keyword table is provided for each genre of music is that, even if the characteristic value of a certain acoustic characteristic is the same, if the genre of the music is different, a keyword suitable for expressing the acoustic characteristic is Because it can be different. The reason why an introduction sentence template is provided for each genre of music is that, if the genre of music is different, a suitable style as an introduction sentence may be different.
As mentioned above, although two examples have been given as embodiments of the music introduction sentence database 152, the embodiment of the music introduction sentence database that the music introduction sentence generation apparatus according to the present invention should have is limited to one of the above two embodiments. Needless to say, it is not something. In short, as long as it is an aspect that can identify (or generate) one introduction sentence data from the characteristic values of each acoustic characteristic obtained by analyzing the music data of the music to be processed, what kind of aspect is it? May be.

  The OS program 153 is a program for causing the control unit 110 to realize an operating system (hereinafter referred to as “OS”). When the liner notes generating apparatus 10 is turned on (not shown), the control unit 110 immediately loads the OS program from the nonvolatile storage unit 150 to the volatile storage unit 140 and starts executing the OS program. The control unit 110 that operates according to the OS program 153 and realizes the OS performs operation control of each unit of the linernotes generating device 10, and performs other control according to a predetermined execution schedule and an instruction of the operation manager. The execution of the program can be started. In the liner notes generating apparatus 10 according to the present embodiment, the control unit 110 operating according to the OS program starts each of the content distribution program 154 and the content registration program 155 as a so-called daemon process.

  When the content distribution program 154 receives a communication message requesting the distribution of digital content, the content distribution program 154 sends the corresponding digital content (in this embodiment, music data and introduction text data corresponding to the music data) to the music database 151. Is a program that causes the control unit 110 to execute a process of reading from the communication message and returning it to the transmission source of the communication message. That is, the content distribution program 154 is a program for causing the liner notes generating apparatus 10 to realize the role of the content server described above.

  The content registration program 155 generates introductory text data indicating an introductory text about the music represented by the music data input via the external device IF unit 130, and associates the introductory text data with the music data to the music database 151. This is a program for causing the control unit 110 to execute the process of registering in. More specifically, the control unit 110 operating according to the content registration program 155 performs the following three processes.

  First, in the feature specifying process of analyzing the music data input via the external device IF unit 130 and specifying the characteristic values indicating each of the plurality of types of acoustic characteristics of the music represented by the music data is there. In the present embodiment, the case where the music data to be processed is input to the liner notes generating apparatus 10 via the external device IF unit 130 has been described, but it may be input via a communication network. When the music data to be processed is input via the communication network, the communication IF unit 120 plays the role of the input means described above.

  Second, a music introduction sentence using a genre identifier (an identifier indicating the genre to which the music belongs) input by the user appropriately operating an operation unit (not shown) and the characteristic value specified by the feature specifying process. This is an introduction sentence acquisition process for searching the database 152 and acquiring corresponding introduction sentence data. In this embodiment, the case where the genre identifier is input via the operation unit will be described. However, it is of course possible to input the genre identifier via the external device IF unit 130. Specifically, a genre identifier is written in association with the music data on a recording medium in which the music data to be processed is written, and the genre identifier is read together with the music data by the control unit 110. good. In the present embodiment, a case where a genre identifier indicating the genre of the music to be processed is input will be described. However, the identification result by the feature specifying process (that is, the degree of each acoustic characteristic of the music to be processed) Of course, the control unit 110 may determine the genre to which the music belongs. In such an aspect, it is not necessary to input a genre identifier.

Thirdly, in association with the music identifier input by the user appropriately operating the operation unit (not shown), the music data of the music to be processed, the introductory text data acquired in the introductory text acquisition process, Is a content registration process for writing the data into the music database 151.
The above is the configuration of the liner notes generating apparatus 10.

(A-2: Operation)
Next, among the operations executed by the liner notes generating device 10, refer to the drawings for the operations that remarkably show the characteristics of the music introduction sentence generating device according to the present invention, that is, the operations performed by the control unit 110 according to the content registration program 155. I will explain. In the operation example described below, the power (not shown) of the liner notes generating apparatus 10 has been turned on, and the content registration program 155 is executed by the control unit 110 as the daemon process described above under the control of the OS. Suppose it has started.

  Under the above circumstances, a recording medium storing the music data to be processed is connected to the external device IF unit 130, and a genre identifier indicating the genre to which the music represented by the music data belongs via an operation unit (not shown). The control unit 110 operating according to the content registration program 155 executes the process shown in the flowchart of FIG. As shown in FIG. 4, the control unit 110 first reads the music data of the music to be processed from the recording medium via the external device IF unit 130, analyzes the music data, and represents the music represented by the music data. Each acoustic characteristic is specified by calculating a characteristic value indicating each of the above-described plural types of acoustic characteristics (acoustic characteristics in which keywords are registered in the music introduction sentence database) (step SA100). In executing step SA100, a known algorithm for each acoustic characteristic may be used. In addition, when specifying the plurality of types of acoustic characteristics for the music to be processed, the degree of each acoustic characteristic may be specified throughout the music, and for each part separated by an interlude such as the first half or the second half of the music The degree of each acoustic characteristic may be specified. In the aspect of specifying the degree of each acoustic characteristic throughout the entire piece of music, one introductory sentence data is generated for the whole piece of music by the following process, and the degree of each acoustic characteristic is specified for each part. In this mode, the following steps SA110 and 120 are executed for each part, and introduction text data is generated for each part.

  In step SA110 executed subsequent to step SA100, the control unit 110 searches the music introduction sentence database 152 using the genre identifier input via the operation unit and the characteristic values specified in step SA100. To obtain introductory sentence data corresponding to the characteristic values and genre identifiers. For example, when the music introduction sentence database 152 is implemented in the first aspect described above, the control unit 110 corresponds to each characteristic value from the introduction sentence table corresponding to the genre identified by the genre identifier. The introductory text data about the music to be processed is acquired by reading the introductory text data to be processed. When the music introduction sentence database 152 is implemented in the above-described second mode, the control unit 110 refers to the keyword table corresponding to the genre identifier and identifies the keyword corresponding to each characteristic value. The introductory sentence data about the music to be processed is acquired by embedding those keywords in the introductory sentence template corresponding to the genre identifier and generating the introductory sentence data.

  Then, the control unit 110 stores the introductory sentence data acquired in step SA110 as liner notes data for the music to be processed in the music database 151 in association with the music identifier for the music and the music identifier of the music. (Step SA120). Thereafter, when receiving the content distribution request message via the communication IF unit 120, the control unit 110 reads content (music data and / or liner notes data) corresponding to the requested content from the music database 151, Deliver to the sender of the content request message. In addition, when performing such content distribution, of course, you may charge according to a predetermined charging algorithm. As described above, the sentence represented by the liner notes data distributed from the liner notes generating apparatus 10 reflects the characteristics of the music paired with the liner notes data. Further, according to the present embodiment, an introductory sentence (liner notes) reflecting the characteristics of the music is automatically generated from the acoustic characteristics of the music and the stored contents of the music introductory text database 152 for any music. Therefore, it doesn't bother the experts such as music critics.

(B: Second embodiment)
Next, a narration adding device 30 according to a second embodiment of the present invention will be described.
The narration adding device 30 is a content server that distributes music and its introduction as digital content, similar to the liner notes generating device 10 described above, and reflects the characteristics of the music by analyzing music data representing the music. The point that the introduction sentence data indicating the introduction sentence is generated is the same as the liner notes generation apparatus 10 described above. However, the narration adding device 30 synthesizes music data in which the narration that is the reading sound of the introductory text indicated by the introductory text data is superimposed on a predetermined section such as a prelude section, an interlude section, and a postlude section of the music. Is different from the liner notes generating apparatus 10 described above.
Hereinafter, the configuration and operation of the narration adding device 30 will be described with reference to the drawings.

(B-1: Configuration)
FIG. 5 is a block diagram showing a configuration example of the narration adding apparatus 30 according to the second embodiment of the present invention. As is apparent from a comparison between FIG. 5 and FIG. 1, the hardware configuration of the narration adding device 30 is the same as the hardware configuration of the liner notes generating device 10 described above (see FIG. 1). Specifically, the narration adding device 30 includes a control unit 110, a communication IF unit 120, an external device IF unit 130, a volatile storage unit 140, a nonvolatile storage unit 150, and a bus 160. However, in order to realize processing characteristic of the narration adding device 30, the software configuration of the narration adding device 30 (data and programs stored in the non-volatile storage unit 150) is the software configuration of the liner notes generating device 10. Is different.

  Specifically, in addition to the music database 151 and the music introduction text database 152, the voice synthesis database 256 is stored in the nonvolatile storage unit 150, and the content registration program 255 is replaced with the content registration program 155. The difference is that it is stored in the storage unit 150. The speech synthesis database 256 is a so-called phonological database, and is a collection of phonological data referred to in speech synthesis. In the speech synthesis database 256, a collection of phoneme data is stored in advance for each of various voice qualities such as a type of generated voice, for example, male voice, female voice, light voice quality, deep voice quality, and the like. The content stored in the speech synthesis database 256 is obtained by converting the text data representing the introduction sentence generated for the music data to be processed into voice data and generating narration data representing the narration for the music represented by the music data. Used for In the present embodiment, a case where the user specifies which voice quality to synthesize is described. For example, a musical piece suitable for narration in a voice synthesized with phonological data in association with phonological data. A genre identifier indicating the genre may be associated, and the narration may be synthesized with a voice quality suitable for the genre to which the music to be processed belongs.

On the other hand, the content registration program 255 searches the music introduction database using each characteristic value obtained by analyzing the music data of the music to be processed and the genre identifier of the music, and reflects the characteristics of the music. The point which makes the control part 110 perform the process which acquires introductory sentence data is the same as the content registration program 155 mentioned above. However, the content registration program 255 synthesizes the narration data indicating the voice read out from the sentence represented by the introductory sentence data acquired as described above from the introductory sentence data and the contents stored in the speech synthesis database, and narrates the voice. As described above, the content registration program 155 is different from the above-described content registration program 155 in that the control unit 110 executes a process of superimposing the music data on the music data.
The above is the configuration of the narration adding device 30.

(B-2: Operation)
Next, among the operations performed by the narration adding device, the operation that remarkably shows the characteristics of the music introduction sentence generating device according to the present invention, that is, the operation executed by the control unit 110 according to the content registration program 255 will be described.
FIG. 6 is a flowchart showing a flow of operations executed by the control unit 110 in accordance with the content registration program 255. As apparent from a comparison between FIG. 4 and FIG. 6, the operation performed by the control unit 110 in accordance with the content registration program 255 is performed by performing three processes from the step SB100 to the step SB120 after executing the process of the step SA110. Only the point that the process of step SA120 is executed later is different from the operation executed by the control unit 110 in accordance with the content registration program 155. Hereinafter, the processing from Step SB100 to Step SB120, which is the difference from the operation of the liner notes generating apparatus 10, will be mainly described.

  In step SB100, the control unit 110 analyzes the music data of the music to be processed, and identifies the section to which narration should be given among the music represented by the music data. For example, when the music data represents a song, the control unit 110 performs spectrum analysis on the music data and changes its spectrum (that is, a spectrum characteristic of a song such as a spectrum characteristic of a human voice). The position of singing is specified from the presence or absence), and the section before the singing position (that is, the prelude section) is specified as the section to be given narration. In this embodiment, the case where the prelude section of the music is used as the section to which narration should be given has been described, but it is needless to say that it may be an interlude section or a follower section. Regarding the specification of the interlude section and the subsequent section, it can be specified by detecting the spectrum characteristic of the singing sound in the same manner as the specification of the prelude section described above. In addition, for music that does not involve singing, the prelude section, etc. may be specified from the change in volume, the part presumed to be the main melody by melody extraction, etc., and the presence of the spectrum belonging to that part, etc. May be specified.

  Next, the control unit 110 generates speech data indicating the speech read out from the introductory sentence represented by the introductory sentence data acquired in step SA110 from the introductory sentence data and phonological data corresponding to the voice quality designated by the user ( In step SB110, the voice data and the music data to be processed are synthesized so that the voice represented by the voice data is superimposed on the narration addition section identified in step SB100 (step SB120). In the present embodiment, the speech of the introductory text represented by the introductory text data obtained by the processing from step SA100 to step SA110 is generated by speech synthesis for the music data to be processed and is superimposed on a predetermined section of the music. Explained. However, introductory text data (for example, introductory text data created by the user himself / herself with a word processor) indicating the introductory text along with the music data to be processed is given to the narration adding device 30, and the processing after step SB100 is performed on the introductory text data. Of course, it may be executed.

  Then, the control unit 110 associates the introductory sentence data generated in step SA110 with the narrated music data synthesized in step SB120 and writes it in the music database (step SA120), and ends this operation. In the present embodiment, a case has been described in which the voice that reads out the entire introduction sentence represented by the introduction sentence data acquired in step SA110 is added as a narration. If speech that reads out the entire sentence cannot be superimposed, speech synthesis may be performed after appropriately summarizing the introduction sentence represented by the introduction sentence data using an existing sentence summarization algorithm. Further, when a plurality of sections are identified as sections to which narration is to be assigned in step SB100, the length of the introduction sentence represented by the introduction sentence data is compared with the time length of each of the plurality of sections. The section to which narration is given may be specified from the section.

  Thereafter, when receiving the content distribution request message via the communication IF unit 120, the control unit 110 reads content (music data and / or liner notes data) corresponding to the requested content from the music database 151, Deliver to the sender of the content request message. As a result, arbitrary music data can be converted into music data representing music obtained by adding a narration reflecting the characteristics of the music to the prelude section of the music represented by the music data, and can be distributed. Also in the second embodiment, it is the same as that in the first embodiment described above that when the music data with narration is generated, an expert such as a music critic is not bothered. .

(C: other embodiment)
Although one embodiment of the present invention has been described above, it goes without saying that modifications described below may be added to such an embodiment.
(1) In each of the above-described embodiments, a case has been described in which processing characteristic of the music introduction sentence generation device according to the present invention is implemented by a software module. At least a part or all of the processing is implemented by an electronic circuit (ie, hardware). Of course, it may be realized by a module).

(2) In the above-described embodiment, the case has been described in which the contents stored in the music introduction sentence database are classified for each genre of music, but such classification is not necessarily essential. For example, in the first embodiment described above, a common introduction sentence table may be used for each genre, and in the second embodiment, a keyword common to each genre. Of course, a common introductory sentence template may be used across tables and genres.

  In the above-described embodiment, the music data of the music to be processed is analyzed, the characteristic values indicating the characteristics of the music for a plurality of predetermined acoustic characteristics are calculated, and the characteristic values are used. The case where introductory text data indicating the introductory text for the music is acquired has been described. For example, the introductory text that indicates the name of the music or the name of the performer that is input by appropriately operating the operation unit Of course, it may be added to the sentence. Specifically, a text template “Please enjoy XX playing XX” is stored in the music introduction database, and the name of the player entered by the user in the XX part of the text template. And the text data obtained by embedding the music name input by the user in the □□ portion may be added to the end of the introduction text data.

(3) In each of the above-described embodiments, a program (content registration program) that causes the liner notes generating device 10 or the narration adding device 30 to execute processing characteristic of the music introduction sentence generating device according to the present invention on the control unit 110. ) Was previously installed. However, the content registration program may be recorded and distributed on a computer-readable recording medium such as a CD-ROM or a DVD (Digital Versatile Disk), and the content may be distributed by downloading via an electric communication line such as the Internet. Registered pros may distribute rams.

It is a block diagram showing an example of composition of liner notes generating device 10 concerning a 1st embodiment of the present invention. It is a figure for demonstrating the function of the music introduction sentence database 152 which the liner notes production | generation apparatus 10 has. It is a figure for demonstrating an example of the storage content of the music introduction sentence database 152 which concerns on a 1st embodiment. It is a flowchart which shows the flow of the liner notes production | generation process which the control part 110 of the liner notes production | generation apparatus 10 performs. It is a block diagram which shows the structural example of the narration addition apparatus 30 which concerns on 2nd Embodiment of this invention. It is a flowchart which shows the flow of the narration provision process which the control part 110 of the narration addition apparatus 30 performs.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 10 ... Liner notes production | generation apparatus, 30 ... Narration addition apparatus, 110 ... Control part, 120 ... Communication IF part, 130 ... External apparatus IF part, 140 ... Volatile memory part, 150 ... Nonvolatile memory part, 151 ... Music database, 152 ... Music introduction text database, 153 ... OS program, 154 ... Content distribution program, 155, 255 ... Content registration program, 256 ... Speech synthesis database, 160 ... Bus.

Claims (6)

A music introduction database that stores data for generating or identifying one music introduction from characteristic values indicating each of a plurality of predetermined acoustic characteristics;
Input means for inputting music data;
Analyzing the music data received by the input means, and specifying a characteristic value indicating each of the plurality of types of acoustic characteristics;
Obtain and output introductory text data representing an introductory text about the music represented by the music data received by the input means from each characteristic value specified by the feature specifying means and the stored contents of the music introductory text database And a music introduction sentence generating device characterized by comprising: The contents stored in the music introduction sentence database are classified for each genre of music,
The acquisition means is a genre identifier indicating a genre to which a music represented by music data received by the input means belongs, and a genre indicated by a genre identifier input from the outside, or each characteristic value specified by the feature specifying means Referencing data stored in the music introduction sentence database in association with the genre of the music specified from the above, obtaining introductory sentence data about the music represented by the music data received by the input means, The music introduction sentence generation apparatus according to claim 1. Input means for inputting music data;
Analyzing the music data received by the input means, and a section specifying means for specifying a prelude section, an interlude section or a postlude section of the music represented by the music data;
Acquisition means for acquiring sentence data representing an introduction sentence reflecting the characteristics of the music represented by the music data received by the input means;
Voice synthesizing means for synthesizing voice data of narration, which is a reading voice of an introduction sentence represented by the sentence data acquired by the acquiring means, from the sentence data;
Synthesizing means for synthesizing and outputting the voice data and the music data so that the voice represented by the voice data synthesized by the voice synthesizing means is superimposed on the section specified by the specifying means. Narration adding device. A music introduction database that stores data for generating or identifying one music introduction from characteristic values indicating each of a plurality of predetermined acoustic characteristics;
Analyzing the music data received by the input means, and specifying a characteristic value indicating each of the plurality of types of acoustic characteristics;
With
The acquisition means acquires introductory sentence data representing an introductory sentence about the music represented by the music data from each characteristic value specified by the feature specifying means and the stored contents of the music introductory text database. 4. The narration adding device according to claim 3, wherein Computer equipment,
Analyzing the music data input to the computer device, and a characteristic specifying means for specifying a characteristic value indicating each of a plurality of predetermined acoustic characteristics;
The stored contents of the music introduction sentence database storing data for generating or specifying one music introduction sentence from the characteristic values indicating each of the plurality of types of acoustic characteristics, and the characteristic values specified by the feature specifying means, From the above, the program is made to function as an acquisition means for acquiring and outputting introductory text data representing an introductory text for the music represented by the music data. Computer equipment,
Analyzing the music data input to the computer device, section specifying means for specifying the prelude section, the interlude section or the postlude section of the music represented by the music data;
Acquisition means for acquiring sentence data representing an introductory sentence reflecting the characteristics of the music represented by the music data;
Voice synthesizing means for synthesizing voice data of narration, which is a reading voice of an introduction sentence represented by the sentence data acquired by the acquiring means, from the sentence data;
Functioning as a synthesizing unit that synthesizes and outputs the voice data and the music data so that the voice represented by the voice data synthesized by the voice synthesizing unit is superimposed on the section specified by the specifying unit. A program characterized by

Images