JPH08137902A - Device for retrieving image space area specifying sound - Google Patents

Abstract

PURPOSE: To enable a general user to easily retrieve a required sound from a sound library with enormous volume by parametrically applying an image having the sound required by the user as a motive for retrieval within a range of the evaluation value of a sound evaluation element. CONSTITUTION: A sound retrieving means 3 compares a range of the evaluation value of an evaluation element inputted by an image parameter input means 2 with data stored in a sound image evaluation data storing part 1 and retrieves a sound within the range. The means 3 stores the sound number, name, etc., of the sound included in sound image evaluation data, the evaluation data of each evaluation element, the storing address of recording information to be reproduced as a sound, etc., in a specific data storing part and displays these contents on a screen as a sound table. When three evaluation elements or less among plural ones are specified in addition to the sound table, sound image evaluation data to be a retrieved result are two-dimensionally or three- dimensinally displayed and each sound can be visually distinguished.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a video system and a presentation system using multimedia, and a harp having a background music (BGM) and a short melody for producing an atmosphere of the scene of the video for a certain length of time. Music effect (ME), which is a sound effect such as scene change made with sounds of chimes, bells, musical instruments, etc., even if the same sound effect is close to that emitted by a real thing, for example, sound of waves, door closing sound, thing The present invention relates to a sound search device for searching various sounds (sounds) added to an image such as a falling sound, a sound effect (SE) such as an explosion sound, and the like.

Due to the high performance of computer systems in recent years, multimedia systems using personal computers are being realized. In such a system, not only characters as in the past but also media such as images, sounds, and moving images can be easily handled, and are used for home videos, animations, and presentations.

Currently, libraries for these systems are being prepared, but the system for searching for a desired object is lagging behind. Therefore, there is a strong demand for a technology that allows even a user who has never heard the sound in the library to search it.

[0004]

2. Description of the Related Art Regarding video, a method of performing a list search by increasing the number of displays per screen by reducing the resolution at the expense of resolution is common.

However, since it is not possible to distinguish between sounds placed spatially in parallel, it is necessary for the user to listen to the sound in his spare time and create an index by himself. Sometimes I used the index to find the desired sound.

As described above, there has not been a conventional technique for systematically searching for a sound. Although some libraries had an index, it was a system that searched using the limited words prepared in advance in the system, and fuzzy search could not be performed.

[0007]

As described above, there is a problem that it is not sufficient to express the image of the user and search for the sound because the fuzzy search cannot be performed.

In view of the above points, the present invention provides means for searching for a sound by giving an image that a user has for a sound as a motivation for the search to one or more image evaluation elements with a range of evaluation values as a condition. The purpose is to provide.

[0009]

The above-mentioned problems can be solved by an image space area designated sound retrieval apparatus constructed as follows.

FIG. 1 is a block diagram of the present invention. The sound image evaluation data storage unit 1 stores a database in which a sound is evaluated according to one or more evaluation factors, and the sound is associated with the evaluation value of each image evaluation factor and the storage address of the recording information for reproducing the sound. The image parameter input means 2 is means for designating the range of the evaluation value for each image evaluation element in the first invention.

The sound search means 3 searches the sound image evaluation data storage section 1 for a sound that is evaluated and exists in the image space area formed by the range designated by the image parameter input means 2.

The sound map display / selection means 4 is additionally provided in the second and third inventions,
In the first invention, the searched sound is displayed so as to be positioned in the three-dimensional or two-dimensional space on the screen for arbitrary three or less evaluation elements, and the sound is visually distinguished and selectively specified.

[0013]

In the first aspect of the present invention, the range of the evaluation value of the evaluation element input by the image parameter input means 2 is compared with the data in the sound image evaluation data storage unit 1,
When a sound within the range is searched for, it is possible to reproduce it as a sound, which is not shown in the figure, but is separately stored in the data storage section, such as the sound number and name in the sound image evaluation data, the evaluation data of each evaluation element, and the like. The storage address and the like of the recording information is accumulated and stored, and a list of sounds is displayed on the screen.

In the second invention, by specifying not more than three of the sound evaluation elements in addition to the sound list displayed on the screen according to the first invention, the image evaluation data of the sound of the search result is specified. It is possible to visually distinguish sounds by displaying them on a two-dimensional or three-dimensional screen.

In the third invention, even when the image space in the range specified in the second invention does not exist, a search is performed by incorporating a candidate having a short distance from the specified range as a candidate.

[0016]

FIG. 2 is a block diagram of an embodiment of the present invention. The computer unit 11 is provided with a unit that realizes various functions under the control of the search control unit 21, and a data storage unit 27 that stores, processes, and delivers data by each unit.

As a device connected to the computer unit 11, a sound image evaluation data storage unit 1 for storing sound image evaluation data, and sound data which is recording information for performing reproduction performance as sounds of each sound. There is a sound data storage unit 14 for storing.

Further, as an interface with a user who searches for a sound, a sound search device is provided with a display / input unit 12 for inputting and outputting through a screen and a sound playing unit 15 for playing the searched sound as a sound. .

FIG. 3 is a search flowchart. The sound search mechanism will be described according to the flowchart.
The search flow is such that the search control means 21 activates means for carrying out each function, and each means and each part of the apparatus will be described according to the flow.

In step S31 [input image parameter], the image parameter input means 2 operates. FIG. 4 shows an explanatory diagram of the image parameter input screen. FIG. 4 shows a case where the user specifies the range of the evaluation value of each evaluation element in the range 44 item on the screen for the two elements of brightness and warmth out of the evaluation elements 43 of the plurality of sounds. There is.

The evaluation value of each evaluation element is from -100 to +10.
It is expressed as a value up to 0. Negative indicates that the degree of having the element is small, and positive indicates that the degree of the element is large. For example, when the brightness is +100, the sound of an extremely bright image is shown, and when the brightness is -100, a dark image has an extremely strong element. The intermediate value indicates the degree of brightness and darkness by a numerical value. These numerical values are the numerical values that statistically many people feel statistically using a questionnaire or the like.

In FIG. 4, the brightness factor is −
It is indicated by bar display 44a that it specifies to search for sounds that are rated in the range 50 to +80. Similarly, the warmth is specified from 0 to +100 on the bar display 44b.

For other evaluation elements 43, bars (44c, 44d, 4
4e, 44f) is displayed from −100 to +100, which does not give a condition for these evaluation elements 43,
Alternatively, it indicates that the entire range is positively specified, that is, the evaluation value of these evaluation elements 43 does not matter.

The display element 42 in the leftmost column in FIG. 4 designates an element whose search result is shown in the sound map, and any three or less can be designated. This is because the sound map (71, 81) is displayed on the screen, and the three-dimensional display is the limit.

Since the display is performed with the evaluation element 43 with the designated * mark as the axis, a two-dimensional screen with the two evaluation elements of brightness and warmth as axes is displayed as the sound map 71. It will be. As will be described later, it is as shown in FIG.

Next, in step S32 [Sound search], the sound image evaluation data storage unit 1 is searched based on the image parameters input as described above.

First, the structure of the sound image evaluation data will be described. FIG. 5 shows an explanatory diagram of the sound image evaluation data. The configuration is such that the evaluation of each of the plurality of evaluation elements 52 corresponds to the sound number and name, and the evaluation is from -100 to +100.
It is a tabular database given as an evaluation value between the two. Sound data address at the right end as an item
53 is recorded and indicates the address where the digital recording information in which the recording data is recorded in order to reproduce each sound as an actual sound is stored.

The sound evaluation elements 52 are distributed so as to distinguish the sounds by using 3 to 6 kinds. The evaluation value for each sound evaluation element is set by statistically processing the degree to which the general public feels by a unified method.

The sound search means 3 searches each sound in the sound image evaluation data storage section 1 for a sound that falls within the range of the evaluation value stored in the data storage section 27 by the image parameter input section 2 for each evaluation element. The search results are stored in the data storage unit 27 as a table in the same format as the sound image evaluation data 51 of FIG.

Since the search is performed by giving evaluation value ranges for some evaluation elements, the number of sound candidates in the search result is large, and there is a strong possibility that it is necessary to narrow down the target sound. How to find the desired sound from these candidates is determined in step S33 [Refine by map? ] To contact the user.

As a method corresponding to the first invention, N is selected here. That is, the process proceeds to [Sound title display] in step S36. FIG. 6 shows an explanatory diagram of the sound title display screen.

The results retrieved as described above are stored as a table in the data storage unit 27, and as shown in FIG. 6, the sound title display means 24 displays this in a format similar to that of FIG. When you cannot fit one screen, you can see the hidden part by scrolling on the screen.

As a result, the user looks at the sound list 61 selected as a candidate, and in the performance selection field 62, a sound that seems to be the target sound is selected from the name of the sound and the evaluation points of the evaluation elements *.
By selecting with a mark, it is possible to cause the sound playing section 15 to play as an actual sound in [Sound playing] in step S37.

Sound data in which the sound playing means 26 knows the sound data address corresponding to the sound number designated by the mark from the table of the search results stored in the data storage unit 27 and stores the sound recording information. The performance of the sound is realized by reading out from the corresponding address of the storage unit 14 and sending it to the sound performance unit 15.

If none of the sound title display screens 61 fulfills the purpose, the [Sound of purpose? ] Then, by moving to step S31 and inputting the correction of the image parameter again, it is possible to change the search range and start over.

Next, as a portion corresponding to the second and third aspects of the invention, the case where Y is selected in step S33 will be described.
FIG. 7 is an explanatory diagram of a sound map display (two-dimensional) example. In [Sound map display] in step S34, the sound map display / selection means 4 displays a screen as shown in FIG.

FIG. 7 shows the result of searching in the range specified for each of the evaluation factors of brightness and warmth in FIG.
, Is a plot of the retrieved sound on a quadratic plane created by two axes of warmth 73b. Lower brightness limit 72
c is -50, the upper limit 72a is +80, the left limit 72d of warmth is 0, and the right limit 72b shows a square frame at the position of +100, which is the range given as an image parameter.

Small squares 75 and large squares 74 are shown as the search results in the evaluation of each axis in this rectangular frame. A small square 75 indicates that there is a sound candidate uniquely determined at this evaluation position, and a large square 74 indicates that there are a plurality of sounds having the same numerical evaluation value on these two axes. This includes the case of a sound having a combination of exactly the same evaluation values and the case where the values of the two evaluation elements are the same but the values of the other evaluation elements are different. That is, since the object mapped in the multidimensional space is viewed through the two-dimensional window, such a phenomenon occurs.

There are two small squares (75a, 75b) surrounded by a dotted frame 76 on the upper left side. This dotted frame 76 indicates that the user has selected the range by designating the two X marks in the figure in [Sound selection] in step S35. That is, in this two-dimensional map display, the one on the upper left side, that is, the one with a high degree of brightness and a low degree of warmth was visually selected.

In this way, it is possible to select the sound at this position while visually observing the sound distribution while the title of the sound is not yet displayed. You can visually select and select.

When selecting the sound displayed by the large square 74, it is possible to know in advance that there are a plurality of sounds. The titles of the plurality of sound candidates selected here are displayed in the next step S36, and the following flow is the same as the operation corresponding to the first invention.

The function corresponding to the third invention is a function when the sound is designated by a dotted frame 76 in the sound map of FIG. That is, in such a case, a close one is selected in the vicinity of the point, and the operation when pointing to a single sound is simplified. [Other Embodiments] FIG. 8 is an explanatory view of an example of a sound map display (three-dimensional).

Regarding the sound map display, the display in the two-dimensional display as shown in FIG. 7 has been described in the above description, but it corresponds to the spread of the three-dimensional display due to the recent progress of computer graphics technology. Then, it is used as a more intuitive display method as a means for displaying three-dimensional mapping information.

Although it may be possible in the future to realize true three-dimensional display using holography, in reality, ordinary two-dimensional screen display is used to perform pseudo three-dimensional display. In FIG. 8, the two sounds indicated by the white circles 85 and the black circles 86 are shown mapped to a three-dimensional space with three evaluation elements as axes 82.

The two-dimensional positions (85a, 85b, 85c, 86a, 86b, 86c) projected on the plane formed by each two elements are also displayed. These sounds are selected by creating a space with a rectangular parallelepiped 84 shown in the middle of the figure and indicating its range.

The position and size are designated by using a normal pointer on the screen to show two points on each axis and a total of six points. The projection 83 on the plane formed by the axes is turned on by the user's designation,
Turn off. It is also possible to move the observation position and angle of the screen vertically and horizontally to perform stereoscopic display from another viewpoint.

[0047]

As is apparent from the above description, according to the present invention, not only is a sound searched by name,
By providing a parametric image of the sound desired by the user as a motivation for search within the evaluation value range of the sound evaluation factors, a general user can generate a vast amount of sound library without sound knowledge. By providing a means to easily search for sound from the site, the time, economy, and quality of saving the time required to create a video work, reducing the learning load of sound search technology, improving the appeal of presentations, etc. Has a significant industrial effect.

[Brief description of drawings]

FIG. 1 is a block diagram of the present invention

FIG. 2 is a configuration diagram of an embodiment of the present invention.

[Figure 3] Search flowchart

FIG. 4 is an explanatory diagram of an image parameter input screen.

FIG. 5 is an explanatory diagram of sound image evaluation data.

[Figure 6] Illustration of the sound title display screen

FIG. 7 is an explanatory diagram of a sound map display (two-dimensional) example.

FIG. 8 is an explanatory diagram of a sound map display (three-dimensional) example.

[Explanation of symbols]

 1 sound image evaluation data storage unit 2 image parameter input unit 3 sound search unit 4 sound map display / selection unit 11 computer unit 12 display / input unit 14 sound data storage unit 15 sound playing unit 21 search control unit 24 sound title display unit 26 Sound playing means 27 Data storage

Claims (3)

[Claims] 1. A sound image evaluation data storage unit storing sounds evaluated by one or more image evaluation elements, image parameter input means for designating a range of evaluation values for each image evaluation element, and sound image evaluation data storage. An image space area-specified sound search device, comprising: a sound search means for searching for a sound that is evaluated and exists in an image space area formed by the above-mentioned specified range for a part. 2. A sound map display / selection unit for displaying and selecting sounds in the sound image evaluation data storage unit in a space of three or less specified evaluation elements is additionally provided, and the spatial area is visually specified. The sound retrieval apparatus according to claim 1, wherein the sound in the sound image evaluation data storage section is searched for by doing so. 3. The sound retrieving device for designating an image space region according to claim 2, wherein when the space region does not have a size, a sound in the sound image evaluation data storage unit existing in a nearby space is searched. .