Video distribution device, video distribution method, and video distribution program

Abstract

The present invention makes it possible to output a comment that is likely to attract the interest and attention of a viewer for a gaming video. An attribute information acquisition unit (140) obtains viewer attribute information indicating an attribute of a viewer accessing a video distribution device (10). A comment information selection unit (170) selects at least one of multiple sets of comment information associated with predetermined scenes in a video on the basis of the viewer attribute information. A comment output unit (200) outputs a comment based on the selected set of comment information together with a predetermined scene in the video.

Description

[Technical field]

[0001] The present disclosure relates to a video distribution device, a video distribution method, and a video distribution program. [Background technology]

[0002] Video sharing services via networks such as the Internet are known. In video sharing services, videos posted by users on a server on the network are made available for other users to view. Such videos include, for example, game videos that are recorded while playing a game.

[0003] The following Patent Document 1 describes how, by associating information that identifies a game event with the playback time of a game video, it is possible to output information related to the event when the playback time of the game video arrives. The following Patent Document 1 also describes how a comment corresponding to the event can be output. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent Publication No. 2021-007052 Summary of the Invention [Problem to be solved by the invention]

[0005] For example, if comments corresponding to game events are mechanically associated with game videos, the contents of the comments may not necessarily attract the interest or attention of users (viewers) watching the game videos, even if the comments are output during the playback of the game videos. In this case, there is a concern that the interest or attention of viewers in the game videos themselves may wane.

[0006] An object of the present disclosure is to make it possible to output comments that are likely to attract viewers' interest in game videos. [Means for solving the problem]

[0007] The video distribution device according to the present disclosure is a video distribution device that distributes videos to viewers, and includes an attribute information acquisition means for acquiring viewer attribute information indicating attributes of a viewer who has accessed the video distribution device, a comment information selection means for selecting at least one of a plurality of pieces of comment information associated with a predetermined scene included in the video based on the viewer attribute information, and a comment output means for outputting a comment based on the comment information selected by the comment information selection means together with the predetermined scene included in the video. This makes it possible to output comments that are likely to attract the viewer's interest in the game video.

[0008] The video distribution method according to the present disclosure includes an attribute information acquisition step in which an attribute information acquisition means acquires viewer attribute information indicating attributes of a viewer who has accessed a video distribution device, a comment information selection step in which a comment information selection means selects at least one of a plurality of pieces of comment information associated with a predetermined scene included in the video based on the viewer attribute information, and a comment output step in which a comment output means outputs a comment based on the comment information selected in the comment information selection step together with the predetermined scene included in the video. This makes it possible to output comments that are likely to attract the viewer's interest in the game video.

[0009] A video distribution program according to the present disclosure is a video distribution program that causes a computer to function as a video distribution device that distributes videos to viewers, and causes the computer to function as attribute information acquisition means for acquiring viewer attribute information indicating attributes of a viewer who has accessed the computer, comment information selection means for selecting at least one of a plurality of pieces of comment information associated with a predetermined scene included in the video based on the viewer attribute information, and comment output means for outputting a comment based on the comment information selected by the comment information selection means together with the predetermined scene included in the video. This makes it possible to output comments that are likely to attract the viewer's interest in the game video using a computer. [Brief description of the drawings]

[0010] [Figure 1] 1 is a diagram illustrating a configuration of a system including a video distribution device according to an embodiment of the present disclosure. [Figure 2A] FIG. 2 is a diagram illustrating an example of a hardware configuration of a video distribution device. [Figure 2B] FIG. 13 is a diagram illustrating another example of a hardware configuration of a video distribution device. [Diagram 3] FIG. 2 is a functional block diagram showing an example of functions implemented in the video distribution device. [Figure 4] FIG. 11 is a diagram showing an example of related information related to a video. [Diagram 5] 4 is a diagram showing an example of attribute information acquired by an attribute information acquisition unit; FIG. [Figure 6] 11 is a diagram showing an example of common attribute information identified by a common attribute information acquisition unit; FIG. [Figure 7] 11 is a diagram showing an example of a plurality of pieces of comment information acquired by a comment information acquisition unit; FIG. [Figure 8] FIG. 11 is a diagram showing an example of determination condition information indicating a determination condition for a degree of interest. [Figure 9] 13 is a diagram showing an example of a determination result by an interest level determination unit. FIG. [Figure 10] FIG. 13 is a diagram showing an example of a comment output to a viewer. [Figure 11] FIG. 11 is a diagram showing another example of a comment output to a viewer. [Figure 12] FIG. 11 is a diagram showing another example of a comment output to a viewer. [Figure 13] FIG. 2 is a diagram showing an example of a flow of processing performed by a video distribution device. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] Hereinafter, an embodiment of the present disclosure will be described with reference to the drawings. A video distribution device according to this embodiment outputs comments corresponding to game events along with game videos (hereinafter simply referred to as videos). In particular, the video distribution device according to this embodiment is designed to be able to output comments that are likely to attract the interest and attention of viewers of the video, as described below. Furthermore, the video distribution device is designed to be able to output comments that are likely to attract the interest and attention not only of the video, but also of the distributor of the video.

[0012] [1. System configuration] Fig. 1 is a diagram showing a configuration of a system including a video distribution device 10 according to an embodiment of the present disclosure. As shown in Fig. 1, the video distribution device 10 is connected to a network N such as the Internet so as to be able to transmit and receive data to and from a game server 20, a social networking service (SNS) device 30, and a user terminal 40 via the network N. Since a plurality of game servers 20 exist according to the type of game, a plurality of game servers 20 are connected to the network N. Furthermore, since a plurality of SNS devices 30 exist according to the type of service or community, a plurality of SNS devices 30 are connected to the network N.

[0013] Furthermore, a plurality of user terminals 40 are connected to the network N. Each user using a respective one of the plurality of user terminals 40 is able to view a video transmitted (distributed) to the terminal device 40 by the video distribution device 10.

[0014] A plurality of user terminals 40 are connected to the network N. Each user of the plurality of user terminals 40 receives a video from the video distribution device 10. User terminal It is possible to watch videos sent (distributed) to 40.

[0015] FIG. 2B is a diagram showing another example of the hardware configuration of the video distribution device 10. The video distribution device 10 may be, for example, a game device such as a PlayStation (registered trademark), a personal computer, or a mobile information terminal such as a smartphone or a tablet terminal. In this case, the video distribution device 10 may include a display unit 14 and an operation unit 15 as shown in FIG. 2B. The display unit 14 is a display device such as a liquid crystal display, and displays various images according to instructions from the processor 11. The operation unit 15 is a user interface such as a keyboard, a mouse, or a game controller, and receives an operation input from a user and outputs a signal indicating the content of the operation to the processor 11. The video distribution device 10 may include an optical disk drive that reads an optical disk, a USB (Universal Serial Bus) port, etc.

[0016] The game server 20 is a computer such as a server device, and may include a processor, a storage unit, and a communication unit. The game server 20 is for providing a game to a user, and executes processing of the game played by the user of the user terminal 40. The user terminal 40 communicates with the game server 20, so that the user can play the game provided by the game server 20 via the user terminal 40. The game server 20 may generate a video recording the content of the game play, and store the video in the game database 210 (see FIG. 3) of the game server 20. The game server 20 may also store attribute information indicating the attributes of the user who played the game in the game database 210. The game database 210 may be realized by a storage unit of the game server 20 or a storage device connected to the game server 20.

[0017] The SNS device 30 is a computer such as a server device, and may include a processor, a storage unit, and a communication unit. The SNS device 30 provides a service that promotes interaction between a plurality of users, and provides functions such as posting comments and blog articles, and an electronic bulletin board. The SNS device 30 manages posting information indicating the contents of posts sent by users, information indicating the attributes of users, and the like. The SNS device 30 may store attribute information indicating the attributes of users in an SNS database 310 (see FIG. 3). The SNS database 310 may be realized by a storage unit of the SNS device 30, a storage device connected to the SNS device 30, or the like.

[0018] The user terminal 40 connected to the network N is a computer such as a game device, a personal computer, or a mobile information terminal, and may include a processor, a storage unit, a communication unit, a display unit, and an operation unit. The user terminal 40 may have an audio output unit such as a speaker. A viewer who is a user of the user terminal 40 can watch videos distributed by the video distribution device 10 using the display unit of the user terminal 40 or a display device connected to the user terminal 40.

[0019] [2. Functional blocks] Fig. 3 is a functional block diagram showing an example of functions implemented in the video distribution device 10. As shown in Fig. 3, the video distribution device 10 functionally includes a video information acquisition unit 110, an attribute information collection unit 120, an attribute information storage unit 130, an attribute information acquisition unit 140, a common attribute information identification unit 150, a comment information acquisition unit 160, a comment information selection unit 170, an interest level determination unit 180, an emotional level determination unit 190, and a comment output unit 200. The video information acquisition unit 110, the attribute information collection unit 120, the attribute information acquisition unit 140, the common attribute information identification unit 150, the comment information acquisition unit 160, the comment information selection unit 170, the interest level determination unit 180, the emotional level determination unit 190, and the comment output unit 200 may be realized mainly by the processor 11. The attribute information storage unit 130 may be realized mainly by the storage unit 12, but is not limited thereto, and may be realized by a storage device connected by wire or wirelessly to the video distribution device 10. Note that the video distribution device 10 does not need to implement all of the functions shown in Fig. 3, and may implement functions other than the functions shown in Fig. 3.

[0020] [2-1.Video information acquisition section] The video information acquisition unit 110 acquires related information about a video along with the video (video data). The video includes information such as video, audio, and the video playback time from the start to the end of the video. The related information includes identification information for identifying a specific scene in the video, associated with the video playback time. In this embodiment, the video is a recording of a gameplay by a distributor, and the related information includes event information indicating a specific event that occurred in the gameplay by the distributor, as identification information for identifying a specific scene in the video.

[0021] For example, the video and related information are generated by game play operated by a distributor using the operation unit 15 of the video distribution device 10 or the operation unit of the user terminal 40, and are stored (saved) in the game database 210 of the game server 20. In this case, the video information acquisition unit 110 may acquire the video and its related information stored in the game database 210. In addition, when the video distribution device 10 is a game device, a personal computer, a mobile information terminal, or the like, and the video and related information are stored in the storage unit 12 of the video distribution device 10, the video information acquisition unit 110 may acquire the video and its related information stored in the storage unit 12 of the video distribution device 10.

[0022] Fig. 4 is a diagram showing an example of related information. As shown in Fig. 4, the related information includes a plurality of pieces of event information E1 to E5 as identification information for identifying a plurality of scenes in a video. Each piece of event information E1 to E5 is associated with a video playback time T1 to T5, respectively.

[0023] In the following explanation, the video is a recording of the streamer playing an FPS (First-person shooter) game titled "FPS Game X," and events E1 to E5 are respectively described as the start of a match, a scene in which the character controlled by the streamer is defeated multiple times, a scene in which the character controlled by the streamer searches for items, a scene in which the character controlled by the streamer scores a headshot on an enemy character, and a scene in which the team to which the streamer belongs wins.

[0024] [2-2. Attribute information collection unit, attribute information storage unit] The attribute information collecting unit 120 collects user attribute information from various devices connected via the network N. The attribute information storage unit 130 stores the attribute information collected by the attribute information collecting unit 120. The attribute information collecting unit 120 collects attribute information stored in, for example, the game database 210 of each game server 20 connected to the network N and the SNS database 310 of each SNS device 30 similarly connected to the network N, and stores the collected information in the attribute information storage unit 130.

[0025] [2-3. Attribute information acquisition section] The attribute information acquisition unit 140 acquires viewer attribute information, which is attribute information indicating attributes of a specific viewer who is a user watching a video. The attribute information acquisition unit 140 acquires, for example, viewer attribute information indicating attributes of a viewer who has accessed the video distribution device 10. In this embodiment, the attribute information acquisition unit 140 also acquires distributor attribute information, which is attribute information indicating attributes of a distributor who distributes a video. The attribute information acquisition unit 140 acquires attribute information of viewers and distributors stored in the attribute information storage unit 130.

[0026] Fig. 5 is a diagram showing an example of attribute information acquired by attribute information acquisition unit 140. In the following, it is assumed that "User A" shown in Fig. 5 is a distributor of a video, and "User B", "User C", and "User D" are viewers of the video. As described below, video distribution device 10 is designed to output unique comments for each viewer, User B, User C, and User D, who are viewers of the video.

[0027] The attribute information acquisition unit 140 acquires attribute information (attribute information collected from the game database 210) indicating, for example, game play tendencies, achievements, etc., as viewer attribute information or distributor attribute information. In the example shown in Fig. 5, as distributor attribute information of user A who is a distributor, attribute information indicating that in "FPS game X" played by user A, the game LV (level) is 22, the play time is 18 hours, and the number of times played today is 10, attribute information indicating that in "puzzle game Y" played by user A, the rank is 700, the play time is 900 hours, and attribute information indicating that in "fighting game Z", the number of times cleared is 600, and the play time is 400 hours is acquired.

[0028] 5, the attribute information acquisition unit 140 acquires, as viewer attribute information of user B, a viewer of the video, attribute information indicating that the game level of "FPS game X" played by user B is 10, the play time is 8 hours, and the number of times played today is 1. Also, as viewer attribute information of user C, a viewer, the attribute information acquired indicates that the rank of "puzzle game Y" played by user C is 10. 60 And the play time is 48 In addition, as viewer attribute information for viewer User D, attribute information is obtained that indicates that User D has cleared the game "Fighting Game Z" three times and has played it for five hours.

[0029] In addition, the attribute information acquisition unit 140 acquires attribute information (attribute information collected from the SNS database 310) related to situations, actions, and interests inferred from the contents of posts on the SNS service. In the example shown in Fig. 5, the following attribute information is acquired as distributor attribute information related to user A: attribute information indicating that user A's residence is S prefecture; attribute information indicating that the number of reactions to professional baseball team H (the number of times user A has made positive reactions to posted information related to professional baseball team H) is 22; attribute information indicating that the number of reactions to professional baseball player I is 1; attribute information indicating that the number of reactions to idol R is 24; attribute information indicating that the number of reactions to professional gamer S is 5; and attribute information indicating that user A traveled to O prefecture last week.

[0030] In the example shown in FIG. 5, the attribute information acquisition unit 140 acquires, as the viewer attribute information of user B, attribute information indicating that user B resides in S prefecture, attribute information indicating that the number of reactions to professional baseball team H is 19, and attribute information indicating that the number of reactions to professional baseball player I is 6. In addition, as the viewer attribute information of user C, attribute information indicating that user C resides in T prefecture, attribute information indicating that the number of reactions to professional baseball team L is 8, and attribute information indicating that the number of reactions to professional gamer S is 10. In addition, as the viewer attribute information of user D, attribute information indicating that user D resides in T prefecture, attribute information indicating that the number of reactions to idol R is 120, and attribute information indicating that user D traveled to O prefecture last week (the same as user A's travel destination last week).

[0031] [2-4. Common attribute information identification part] The common attribute information identifying unit 150 identifies, from among the multiple pieces of viewer attribute information acquired by the attribute information acquiring unit 140, viewer attribute information that is common to distributor attribute information indicating the attributes of the distributor of the video, as common attribute information.

[0032] FIG. 6 is a diagram showing an example of common attribute information specified by the common attribute information specifying unit 150. As shown in FIG. 6, the common attribute information specifying unit 150 specifies, for example, attribute information related to "FPS game X" of user B as common attribute information. This is because, as shown in FIG. 5, the viewer attribute information of user B includes attribute information related to "FPS game X", and the distributor attribute information of user A also includes attribute information related to "FPS game X", and these attribute information have a common game title. Similarly, the common attribute information specifying unit 150 specifies, for example, attribute information indicating that user B resides in S prefecture (attribute information common to user A in that user A resides in S prefecture), attribute information indicating the number of reactions to professional baseball team H (attribute information having a team name in common with the professional baseball team to which user A showed a positive reaction), and attribute information indicating the number of reactions to professional baseball player I (attribute information having a name in common with the professional baseball player followed by user A) as common attribute information.

[0033] In the example shown in FIG. 6, the common attribute information identification unit 150 identifies, as common attribute information, attribute information related to "Puzzle Game Y" of User C (attribute information having a title in common with the game played by User A) and attribute information indicating the number of reactions to professional gamer S (attribute information having a name in common with the professional gamer to whom User A gave a positive reaction). Here, attribute information related to "Professional Baseball Team L" of User C may also be identified as common attribute information, since it is common to "Professional Baseball Team H" to which User A gave a positive reaction in that it is a professional baseball team. However, attribute information indicating that User C resides in Prefecture T is not identified as common attribute information, since it has nothing in common with any of the multiple broadcaster attribute information of User A.

[0034] In addition, in the example shown in FIG. 6, the common attribute information identification unit 150 identifies, as common attribute information, attribute information regarding user D's "fighting game Z" (attribute information having the same title as the game played by user A), attribute information indicating the number of reactions to idol R (attribute information having the same name as the idol to which user A showed a positive reaction), and attribute information indicating that last week's travel destination was prefecture O.

[0035] In addition, when multiple pieces of attribute information of a viewer include attribute information indicating that the residence is O prefecture, this attribute information is common in that the travel destination of user A last week, which is indicated by the distributor attribute information of user A, is O prefecture, so the common attribute information identification unit 150 may identify this attribute information as the common attribute information. In this case, the common attribute information identification unit 150 may add information indicating that O prefecture was user A's travel destination last week to the identified common attribute information.

[0036] [2-5. Comment information acquisition section] The comment information acquisition unit 160 acquires multiple pieces of comment information associated with a predetermined scene included in a video. Each of the multiple pieces of comment information acquired by the comment information acquisition unit 160 indicates a candidate comment to be output by the comment output unit 200 (described later) together with a predetermined scene included in a video. The comment information acquisition unit 160 may acquire comment information related to a game from a game database 210 of the game server 20 connected via the network N. The comment information acquisition unit 160 may also acquire comment information stored in the storage unit 12 of the video distribution device 10, another storage device, or the like.

[0037] 7 is a diagram showing an example of a plurality of pieces of comment information acquired by the comment information acquiring unit 160. As shown in FIG. 7, each piece of comment information includes a fixed phrase. For example, in comment information with the content "Today is your [N]th time playing. You're really into it," the character string excluding [N] is the fixed phrase. Also, in comment information with the content "This game is completely different from [puzzle game], can you handle it?", the character string excluding [puzzle game] is the fixed phrase.

[0038] 7, each piece of comment information is associated with one of a plurality of pieces of event information E1 to E5 (see FIG. 4) that indicate a predetermined event that has occurred in the game. In other words, each piece of event information E1 to E5 is associated with a plurality of pieces of comment information.

[0039] In the example shown in FIG. 7, event information E1 indicating the start of a match in a game is associated with comment information such as "Today is the [N]th time you've played. You've really put in the effort," "This game is completely different from a [puzzle game], so can you handle it?", and "I want to fully utilize the hand-to-hand combat skills I've developed in [fighting games]!". Event information E2 indicating a scene in which a character operated by a streamer is defeated many times is associated with comment information such as "I'm surprised you don't get tired of it. You're not good at giving up," "It's not like a [puzzle game] after all," "The [professional gamer] you admire is a faraway presence," and "Maybe you're still in the mood for a [prefecture] trip?". Event information E3 indicating a scene in which a character operated by a streamer scavenges for items is associated with comment information such as "A [professional baseball team] fan, I'll scavenge for items at all costs," "I'm desperate. I'll even scavenge for trash to become a [professional gamer]," and "I'll devour it as if I were scavenged for souvenirs from a [prefecture] trip." Also, event information E4 showing a scene where the character operated by the streamer hits a headshot on an enemy character is associated with comment information such as "It's a hit! A great home run!", "That's a great KO!", "It's a hit! Has this brought me one step closer to [Pro Gamer]?!", and "A headshot dedicated to [Idol]!" Also, event information E5 showing a scene where the streamer's team wins is associated with comment information such as "After our [N]th time, we finally won. I'm so glad," "Maybe it had something in common with [Puzzle Game], but I'm glad we won," and "It's a victory. [Idol] is smiling too."

[0040] [2-6. Comment information selection section, interest level determination section] The comment information selection unit 170 selects at least one of the multiple pieces of comment information acquired by the comment information acquisition unit 160, based on the viewer attribute information acquired by the attribute information acquisition unit 140. That is, the comment information selection unit 170 selects at least one of the multiple pieces of comment information associated with a predetermined scene included in a video, based on the viewer attribute information indicating the attributes of the viewer of the video. In this embodiment, the comment information selection unit 170 selects at least one of the multiple pieces of comment information, based on the viewer attribute information identified as common attribute information by the common attribute information identification unit 150.

[0041] The comment information selection unit 170 selects at least one of the multiple comment information (see FIG. 7) associated with each of the multiple event information E1 to E5 (see FIG. 4) based on the common attribute information of each viewer. For example, based on attribute information related to "FPS game X" among the common attribute information (see FIG. 9) of user B who is a viewer, the comment information selection unit 170 selects comment information with the content "Today is the [N]th time you've played. You're really into it." from the multiple comment information (see FIG. 7) associated with the event information E1, selects comment information with the content "I'm surprised you never get tired of it. You're not good at giving up." from the multiple comment information associated with the event information E2, and selects comment information with the content "Finally, the [N]th time was a charm and you won. I'm so glad." about the FPS game from the multiple comment information associated with the event information E5.

[0042] From the attribute information of user B regarding "FPS game X," it can be inferred that user B has experience playing FPS game X. Therefore, by selecting comment information regarding FPS game X, which is a game played by the distributor in the video, the comment output unit 200, which will be described later, can output comments that are likely to attract the interest and attention of user B.

[0043] Furthermore, with regard to user B, the comment information selection unit 170 selects comment information with the content "Fans of the [professional baseball team] will just hunt around for items" from among the multiple comment information associated with event information E3, and selects comment information with the content "It's a goal! A great home run!" from among the multiple comment information associated with event information E4, based on attribute information regarding the number of reactions by user B to "professional baseball team H." Since it can be inferred from the attribute information regarding the number of reactions by user B to "professional baseball team H" that user B is a baseball fan, by selecting comment information related to professional baseball teams, it becomes possible to output comments that are likely to attract user B's interest and attention.

[0044] Regarding user C, the comment information selection unit 170 selects comment information with the content "This game is completely different from [puzzle game], can you handle it?" from among multiple comment information associated with event information E1, selects comment information with the content "It's not like [puzzle game] after all" from among multiple comment information associated with event information E2, and selects comment information with the content "I wonder if there was something in common with [puzzle game]. Anyway, I'm glad I won" from among multiple comment information associated with event information E5, based on the attribute information regarding the number of reactions of "professional gamer S" of user C, selects comment information with the content "I'm desperate. I'll even dig through trash to become a [professional gamer]" from among multiple comment information associated with event information E3, and selects comment information with the content "It's decided! Is this one step closer to [professional gamer]!?" from among multiple comment information associated with event information E4. From user C's attribute information regarding "puzzle game Y" and the attribute information regarding the number of reactions to "professional gamer S," it can be inferred that user C has experience playing puzzle game Y and is interested in professional gamer S. Therefore, by selecting comment information regarding puzzle game Y and professional gamer S, it is possible to output comments that are likely to attract user C's interest.

[0045] Regarding user D, the comment information selection unit 170 selects comment information with the content "I want to fully demonstrate the hand-to-hand combat skills I have developed in [fighting game]!" from among the multiple comment information associated with event information E1 based on the attribute information related to "Fighting Game Z" of user D. The comment information selection unit 170 selects comment information with the content "Is the [prefecture] travel mood still there?" from among the multiple comment information associated with event information E2 based on attribute information indicating that user D's travel destination last week was O prefecture and attribute information indicating that user A's travel destination last week was O prefecture, and selects comment information with the content "I will devour souvenirs from my [prefecture] trip as if I were rummaging through them" from among the multiple comment information associated with event information E3. Also, based on attribute information regarding the number of reactions of user D to "idol R," comment information with the content "A headshot for [idol]!" is selected from among multiple comment information associated with event information E4, and comment information with the content "It's a victory. [idol] is smiling, too" is selected from among multiple comment information associated with event information E5. From user D's attribute information, it can be inferred that user D has played fighting game Z, has an interest in idol R, and traveled to O prefecture last week. Therefore, by selecting comment information regarding fighting game Z, idol R, and travel destination O prefecture, it becomes possible to output comments that are likely to attract user D's interest.

[0046] The comment information selection unit 170 may determine at least one of a plurality of pieces of viewer attribute information corresponding to a plurality of pieces of comment information associated with a predetermined scene included in a video, based on a plurality of interest levels determined for each of the plurality of pieces of viewer attribute information, and select comment information corresponding to the determined viewer attribute information. In this case, the interest level determination unit 180 determines the viewer's interest level in the attribute indicated by the viewer attribute information. In this embodiment, the interest level determination unit 180 determines the viewer's interest level and the distributor's interest level in the attribute indicated by the viewer attribute information identified as the common attribute information by the common attribute information identification unit 150. Then, the comment information selection unit 170 determines at least one of a plurality of pieces of viewer attribute information corresponding to a plurality of pieces of comment information associated with a predetermined scene included in a video, based on the viewer's interest level and the distributor's interest level determined for each of the plurality of pieces of viewer attribute information, and selects comment information corresponding to the determined viewer attribute information.

[0047] FIG. 8 is a diagram showing an example of determination condition information indicating a determination condition for the interest level. In the example shown in FIG. 8, the interest level is in three stages: "high", "medium", and "low". For example, in an FPS game, it is shown that the interest level is "high" when the game level is 20 or more, the interest level is "medium" when the game level is 5 or more, and the interest level is "low" when the game level is less than 5. The interest level is not limited to three stages, and may be determined in multiple stages other than 3, or may be a numerical value such as a score.

[0048] The interest level determination unit 180 determines the level of interest of each user, who is a viewer, in the attribute indicated by the viewer attribute information (for example, the common attribute information shown in FIG. 6) of each of users B to D who are viewers, and the level of interest of user A, who is a distributor, for each of the viewer attribute information (for example, the common attribute information shown in FIG. 6), based on the determination condition information shown in, for example, FIG. 8. The interest level determination unit 180 may acquire the determination condition information from information stored in the storage unit 12, the game database 210, the SNS database 310, etc., or may generate the determination condition information based on information stored in these storage devices, databases, etc.

[0049] Fig. 9 is a diagram showing an example of a determination result by the interest level determination unit 180. In the example shown in Fig. 9, the overall interest level is determined based on the interest levels of each of users B to D who are viewers and the interest level of user A who is a distributor. The interest level determination unit 180 may determine the overall interest level by giving a predetermined weighting to each of the viewer's interest level and the distributor's interest level (by multiplying a numerical value indicating the interest level by a predetermined coefficient).

[0050] In the example shown in FIG. 5, the game level indicated in the attribute information of the FPS game X of the user B is 10, and the game level indicated in the attribute information of the FPS game X of the user A who is a distributor is 18. Therefore, based on the judgment condition information shown in FIG. 8, the interest levels of the user B and the user A in the FPS game X are both determined to be "medium". Therefore, as shown in FIG. 9, the overall interest levels of the user B and the user A in the FPS game X are also determined to be "medium". In addition, the interest levels of the user B and the user A in the professional baseball team H are both determined to be "high" based on the number of reactions of each user, so the overall interest levels of the user B and the user A in the professional baseball team H are also determined to be "high". In addition, the interest level of the user B in the professional baseball player I is "medium", but the interest level of the user A is "low". In this case, the interest level determination unit 180 may determine the overall interest levels of the user B and the user A in the professional baseball player I to be "low".

[0051] From the play times shown in the attribute information of puzzle game Y of user C and user A shown in FIG. 5, the interest levels of both user C and user A in puzzle game Y are determined to be "high", and as shown in FIG. 9, the overall interest level in puzzle game Y is also determined to be "high". Similarly, the interest levels of user C and user A in professional gamer S are also determined to be "medium". The interest level of user C in professional baseball team L is determined to be "medium", but the attribute information of user A does not include attribute information on professional baseball team L itself, although attribute information on professional baseball team H, which is the same professional baseball team, is present. In this case, the interest level determination unit 180 may determine the overall interest levels of user B and user A in professional baseball player I to be "low".

[0052] 9, the overall interest level of users D and A in fighting game Z is determined to be "low," and the overall interest level of users A in idol R is determined to be "high." Note that, since the determination condition information shown in FIG. 8 does not include a determination condition for the interest level in each user's place of residence and last week's travel destination, a default value ("medium" in the example shown in FIG. 9) may be set for the interest level in the place of residence and last week's travel destination.

[0053] The comment information selection unit 170 may determine at least one of a plurality of pieces of viewer attribute information corresponding to a plurality of pieces of comment information associated with a predetermined scene included in a video based on the overall interest level determined by the interest level determination unit 180 as described above. In the example shown in FIG. 7, the plurality of pieces of comment information associated with the scene of the event information E2 correspond to an FPS game, a puzzle game, a professional gamer, and a travel destination, respectively. Among the plurality of pieces of attribute information of the user C related to these, as shown in FIG. 9, the overall interest level of the "puzzle game Y" is "high", which is the highest compared to the overall interest level of the "professional gamer S", such as "medium". Therefore, the comment information selection unit 170 determines the attribute information related to the puzzle game Y as the attribute information of the user C corresponding to the scene of E2. Then, the comment information corresponding to the attribute information thus determined is selected. That is, based on the attribute information of user C, comment information on a puzzle game with the content "After all, it's not as good as a [puzzle game]," is selected from among multiple pieces of comment information associated with event information E2, rather than comment information on a professional gamer with the content "The [professional gamer] I admire is a faraway presence." In this way, comment information is obtained based on the attribute information of puzzle game Y, which is determined to have the highest level of interest among the attribute information of user C, so that the comment output unit 200, which will be described later, can output comments that are likely to attract user C's interest and attention more effectively.

[0054] [2-7. Emotion level determination section, comment output section] The comment output unit 200 outputs a comment based on the comment information selected by the comment information selection unit 170 based on the attribute information of the viewer to the user terminal 40 of the viewer together with a predetermined scene included in the video.

[0055] The comment output unit 200 outputs the comment together with a predetermined scene included in the video by, for example, displaying text such as a caption indicating the content of the comment on the display unit of the user terminal 40. The text may be displayed outside the display area in which the video is displayed, or may be displayed superimposed on the video. In addition, if the user terminal 40 has an audio output unit such as a speaker, the comment output unit 200 may convert the comment into audio by a predetermined process such as text to speech, and output it to the audio output unit (such as a speaker) of the user terminal 40.

[0056] The comment information includes a fixed phrase. The comment output unit 200 generates a comment including the fixed phrase included in the comment information and a phrase of an attribute indicated by the viewer attribute information, and outputs the comment to the user terminal 40. The comment output unit 200 may generate a comment to which a phrase included in the viewer attribute information, the distributor attribute information, or the like is added. Furthermore, the comment output unit 200 may output the generated comment to the user terminal 40 of the distributor of the video in order to confirm the content of the generated comment with the distributor of the video, or may display the content of the generated comment on the display unit 14 of the video distribution device 10, or the like.

[0057] 10 to 12 are diagrams showing examples of comments output to viewers. FIG. 10 shows a comment output for user B. As shown in FIG. 10, the comment output unit 200 outputs a comment for user B to the user terminal 40 used by user B. For example, by replacing [N] in "Today is the [N]th play. You're really into it." with "10" included in the distributor attribute information for FPS game X of user A, a comment with the content "Today is the 10th play. You're really into it." is generated and output in association with a scene corresponding to event information E1. Also, a comment with the content "You never get tired of it. You're not good at giving up." is output in association with a scene corresponding to event information E2. Also, by replacing [professional baseball team] in "A [professional baseball team] fan, I'll just hunt for items" with professional baseball team H, which is an attribute indicated by the attribute information of user B, a comment with the content "A professional baseball team H fan, I'll just hunt for items." is generated and output in association with a scene corresponding to event information E3. Similarly, comments such as "That's it! A great home run!" and "The 10th time's the charm, we finally won. I'm so glad," are generated and output in association with the scenes corresponding to the event information E4 and E5, respectively.

[0058] As a result, in the user terminal 40 that played the video, the video playback time T1 to T 5 When each of these comments arrives, comments such as "Today is your 10th time playing. You're really into it," "I'm surprised you don't get tired of it. You never give up," "Professional baseball Team H fans, just rummage through items," "There it is! A great home run!", and "The 10th time's the charm, and you finally won. I'm so glad." are output in sequence. This makes it possible to output comments that are more likely to attract the interest and attention of User B, who has experience with FPS game X, the content of the video, and who has shown a positive reaction to Professional Baseball Team H.

[0059] FIG. 11 shows a comment output for user C. FIG. 12 shows a comment output for user D. As shown in FIG. 11, the comment output unit 200 generates a comment for user C by adding attribute words (puzzle game Y, professional gamer S) indicated by attribute information of user C to a fixed phrase included in the comment information, and outputs the comment to the user terminal 40 of user C. Similarly, as shown in FIG. 12, a comment for user D by adding attribute words (fighting game Z, idol R, O prefecture) indicated by attribute information of user D to a fixed phrase included in the comment information, and outputs the comment to the user terminal 40 of user D. This makes it possible to output comments that are more likely to attract interest and attention to users C and D.

[0060] In this embodiment, comment information is selected based on not only the attribute information of users B to D who are viewers, but also on common attribute information among these attribute information that is common to the attribute information of user A who is a distributor, and the comment of the comment information is output together with each scene of the video. In other words, the comments output to each of users B to D are output based not only on the viewer attribute information of each user, but also on the attribute information of user A who is a distributor. Therefore, it becomes possible to output comments that are likely to attract interest and attention not only to the video but also to user A who is a distributor of the video.

[0061] The comment output unit 200 may output a comment based on the comment information in a mode according to the emotionality of the comment information determined by the emotionality determination unit 190. In this case, the emotionality determination unit 190 may determine the emotionality of the comment information based on the degree of interest determined by the interest level determination unit 180 for the viewer attribute information corresponding to the comment information.

[0062] For example, when the degree of interest determined for the viewer attribute information corresponding to the comment information is determined to be high, the emotionality determination unit 190 determines to increase the emotionality of the comment information, and when the degree of interest is determined to be low, the emotionality determination unit 190 determines to decrease the emotionality of the comment information. For example, comment information with the content "It's a hit! A great home run!" is selected based on the attribute information of user B regarding baseball team H. Here, as shown in FIG. 9, the interest level for professional baseball team H is determined to be "high," so the emotionality determination unit 190 determines to increase the emotionality of comment information with the content "It's a hit! A great home run!".

[0063] For example, when outputting comments to viewers by displaying text such as captions indicating the contents of the comments on the display unit of the user terminal 40, the comment output unit 200 may display comments (text such as captions) based on comment information for which it has been determined that the emotionality level is to be high larger than other comments, or may display them in a different font, color, or the like from the other comments. Also, the comment output unit 200 may display comments based on comment information for which it has been determined that the emotionality level is low smaller than other comments, or may display them with lower brightness or saturation than the other comments.

[0064] Furthermore, when converting comments into voice by a predetermined process such as text to speech, the comment output unit 200 may make the voice of a comment based on comment information determined to have a high emotionality louder than the voices of other comments, or may make the voice more emotional. Furthermore, the comment output unit 200 may make the voice of a comment based on comment information determined to have a low emotionality quieter than the voices of other comments, or may make the voice less emotional.

[0065] As described above, the emotionality determination unit 190 determines the emotionality of the comment information based on the interest level of the viewer attribute information, and the comment output unit 200 outputs the comment in a manner according to the emotionality of the comment information, so that the comment corresponding to the viewer attribute information determined to have a higher interest level is emphasized and output. This makes it possible to output comments that are more likely to attract the viewer's interest and attention more effectively.

[0066] [3. Flowchart] Fig. 13 is a diagram showing an example of the flow of processing performed in video distribution device 10. Processor 11 of video distribution device 10 executes the processing shown in Fig. 13 when, for example, a predetermined viewer (e.g., any one of users B to C) accesses video distribution device 10 using user terminal 40. Processor 11 of video distribution device 10 may also execute one or more of steps S101 to S105 of the processing shown in Fig. 13 when a video distributor (e.g., user A) instructs the predetermined viewer to output the video, and then executes the remaining processing when the viewer accesses video distribution device 10.

[0067] 13, first, the attribute information acquisition unit 140 acquires a plurality of pieces of distributor attribute information related to distributors of videos (step S101). The attribute information acquisition unit 140 acquires attribute information related to the distributors as distributor attribute information from the attribute information collected in advance by the attribute information collection unit 120 from the game database 210, the SNS database 310, etc.

[0068] When the distributor of the video is user A, in step S101, the attribute information acquisition unit 140 acquires attribute information of user A as distributor attribute information. For example, as shown in Fig. 5, attribute information on FPS game X, puzzle game Y, and fighting game Z played by user A, attribute information indicating that user A resides in S prefecture, attribute information on professional baseball team H, professional baseball player I, idol R, and professional gamer S to which user A showed positive reactions, and attribute information indicating that user A traveled to O prefecture last week are acquired as distributor attribute information.

[0069] Next, the attribute information acquiring unit 140 acquires a plurality of pieces of viewer attribute information related to the predetermined viewer (step S102). The attribute information acquiring unit 140 acquires, as the viewer attribute information, attribute information related to the predetermined viewer from among attribute information previously collected from the game database 210, the SNS database 310, etc. Note that the process of step S102 may be executed before the process of step S101.

[0070] If the predetermined viewer is user B, in step S102, the attribute information acquisition unit 140 acquires attribute information of user B as viewer attribute information. For example, as shown in Fig. 5, attribute information on FPS game X played by user B, attribute information indicating that user B's residence is S prefecture, and attribute information on professional baseball team H and professional baseball player I to which user B showed a positive reaction are acquired as viewer attribute information.

[0071] Next, the common attribute information identifying unit 150 identifies common attribute information common to any of the plurality of pieces of viewer attribute information acquired in step S101 from the plurality of pieces of viewer attribute information acquired in step S102 (step S103).

[0072] 6, viewer attribute information regarding FPS game X, residence prefecture S, professional baseball team H, and professional baseball player I of user B are all common to the broadcaster attribute information of user A. Therefore, in step S103, the common attribute information identification unit 150 identifies these viewer attribute information of user B as common attribute information.

[0073] Next, the comment information selection unit 170 selects, for each scene in the video, one of the multiple pieces of comment information associated with that scene based on the viewer attribute information acquired in step S102 (in this embodiment, the common attribute information identified in step S103) (step S104). The comment information selection unit 170 selects one of the multiple pieces of comment information acquired from the game database 210 by the comment information acquisition unit 160 based on the common attribute information.

[0074] For example, as shown in FIG. 7, event information E1 to E2 each indicating a specific event of a game played by a distributor, which is a specific scene included in a video, 5 In step S104, the comment information selection unit 170 selects one of the multiple comment information associated with the event information E1 based on the common attribute information identified in step S103. 5 For each of the event information, one of the multiple comment information associated with each event information is selected based on the common attribute information identified in step S103.

[0075] In step S104, the comment information selection unit 170 selects the event information (event information E1 to E 5 The interest level determination unit 180 may determine multiple pieces of viewer attribute information corresponding to multiple pieces of comment information associated with the multiple pieces of viewer attribute information (any of the above), based on the level of interest determined for each of the multiple pieces of viewer attribute information by the interest level determination unit 180, and select the comment information corresponding to the determined viewer attribute information.

[0076] For example, the interest level determination unit 180 determines the level of interest shown in Fig. 9 as "high", "medium", or "low" based on the determination condition information shown in Fig. 8. The interest level determination unit 180 may determine the level of interest in the viewer attribute information (common attribute information) based on not only information (parameters) such as the game level and play time shown in the viewer attribute information, but also information such as the level and play time shown in the distributor attribute information.

[0077] In step S104, the comment information selection unit 170 may preferentially select comment information corresponding to viewer attribute information determined to have a higher interest level. In the example shown in Fig. 9, the interest level of professional baseball team H among the viewer attribute information of user B is determined to be "high", so the comment information selection unit 170 preferentially selects comment information related to professional baseball teams.

[0078] Next, the comment output unit 200 generates a comment based on the comment information selected in step S104 (step S105). The comment output unit 200 generates a comment including a template included in the comment information and a phrase of an attribute indicated by the viewer attribute information. Then, the comment output unit 200 outputs the comment generated in step S105 to the user terminal 40 (step S106), and ends the process.

[0079] In step S104, for example, if comment information with the content "Professional baseball team fan, just rummage through items," is selected based on user B's attribute information regarding professional baseball team H, in step S105, the comment output unit 200 replaces the [professional baseball team] included in this comment information with "Professional baseball team H," thereby generating a comment with the content "Professional baseball team H fan, just rummage through items," as shown in FIG. 10.

[0080] Also, in step S105, the comment output unit 200 may generate a comment including a phrase indicated in the viewer attribute information or the distributor attribute information. For example, by replacing [N] in the comment information with the content "This is your [N]th play today. You're really into it," with "10," which is the number of plays today, included in the distributor attribute information of user A shown in FIG. 5, it is possible to generate a comment with the content "This is your 10th play today. You're really into it."

[0081] In step S106, the comment output unit 200 outputs the comment together with a predetermined scene included in the video by displaying text such as a caption indicating the content of the comment on the display unit of the user terminal 40. Also, in step S106, the comment output unit 200 may convert the comment into audio by a predetermined process such as text to speech, and output the audio output unit of the user terminal 40.

[0082] Furthermore, in step S104, the emotional level determination unit 190 may determine the emotional level of the comment information based on the interest level of the viewer attribute information. For example, when it is determined that the interest level in the viewer attribute information corresponding to the comment information is high, the emotional level determination unit 190 determines to increase the emotional level of the comment information, and when it is determined that the interest level is low, the emotional level determination unit 190 determines to decrease the emotional level of the comment information.

[0083] In step S106, the comment output unit 200 may output the comment based on the comment information in a mode according to the emotionality of the comment information determined in step S104. The comment output unit 200 may display the comment (text) based on the comment information determined to have a high emotionality larger than the other comments, or may display the comment in a different font, color, or the like from the other comments. Furthermore, the comment output unit 200 may output the voice of the comment based on the comment information determined to have a high emotionality larger than the voice of the other comments.

[0084] In the example shown in FIG. 9, the interest level of the professional baseball team H in the viewer attribute information of the user B is determined to be "high", so in step S104, the emotionality determination unit 190 determines that the emotionality of the comment information related to the professional baseball team is high. In step S106, the comment output unit 200 outputs comments based on the comment information related to the professional baseball team shown in FIG. 10, such as "Professional baseball team H fans, just hunt for items" and "It's a home run! What a home run!", in a more emphasized manner than other comments. In this way, the comments determined to have a high interest level are output in a more emphasized manner than other comments, so that comments that are more likely to attract the viewer's interest in the video can be output.

[0085] [4. Summary] As described above, in this embodiment, the comment information selection unit 170 selects at least one of a plurality of pieces of comment information associated with a predetermined scene included in a video based on viewer attribute information indicating the attributes of the viewer, and the comment output unit 200 outputs a comment based on the comment information selected by the comment information selection unit 170 together with the predetermined scene included in the video to the user terminal 40. In this way, it is possible to output a comment that is likely to attract the viewer's interest and attention in the video.

[0086] In this embodiment, the comment information selection unit 170 selects at least one of the multiple comment information associated with a predetermined scene based on the viewer attribute information identified as common attribute information common to the distributor attribute information indicating the attribute of the distributor of the video. It is possible to output comments that are likely to attract the viewer's interest and attention not only to the video but also to the poster of the video.

[0087] In this embodiment, the comment information includes a fixed phrase, and the comment output unit 200 outputs a comment including the fixed phrase included in the comment information and a phrase of an attribute indicated by the viewer attribute information. In this way, the phrase of an attribute indicated by the viewer attribute information can be reflected in the content of the comment, and a comment that is likely to attract the viewer's interest and attention more effectively can be output.

[0088] Furthermore, in this embodiment, the comment information selection unit 170 determines at least one of a plurality of pieces of viewer attribute information (common attribute information) corresponding to each of a plurality of pieces of comment information associated with a predetermined scene based on a plurality of interest levels determined for each of these pieces of viewer attribute information, and selects comment information corresponding to the determined viewer attribute information. In this manner, comment information corresponding to viewer attribute information determined to have a high interest level is preferentially selected, so that comments that are likely to more effectively attract the viewer's interest in the video can be output.

[0089] In this embodiment, the emotionality determination unit 190 determines the emotionality of the comment information based on the interest level determined for the viewer attribute information corresponding to the comment information, and the comment output unit 200 outputs the comment based on the comment information in a manner according to the determined emotionality. In this manner, for example, a comment determined to have a high interest level can be output in a manner that emphasizes it more than other comments, so that a comment that is likely to more effectively attract the viewer's interest in the video can be output.

[0090] [5. Modifications] The present invention is not limited to the above embodiments.

[0091] (1) In the embodiment, the common attribute information identification unit 150 identifies viewer attribute information common to distributor attribute information indicating the attribute of the distributor of the video among the plurality of viewer attribute information acquired by the attribute information acquisition unit 140 as the common attribute information, and the comment information selection unit 170 selects at least one of the plurality of comment information associated with a predetermined scene based on the common attribute information. Not limited to this, the comment information selection unit 170 may select comment information based on viewer attribute information common to the distributor attribute information and viewer attribute information that is not common. In this case, the video distribution device 10 does not necessarily have to have the common attribute information identification unit 150. Also, the attribute information acquisition unit 140 does not need to acquire poster attribute information, and the interest level determination unit 180 may determine the interest level of only the viewer in the attribute indicated by the viewer attribute information. Even in this way, it is possible to output comments that are likely to attract the viewer's interest in the video.

[0092] (2) In the embodiment, an example has been described in which a comment is generated and output together with a predetermined scene contained in a video when a predetermined viewer (e.g., any one of users B to C) accesses video distribution device 10 using user terminal 40, or when a video poster instructs a predetermined viewer to output the video. This method is suitable for when a distributor distributes a video that has been recorded in advance (recorded distribution), but the present invention can also be applied when a distributor distributes a video in real time (live distribution).

[0093] In this case, the video information acquisition unit 110 (see FIG. 3) acquires in real time the video recorded in the game server 20, the user terminal 40, etc., and related information related to the video. The attribute information acquisition unit 140 acquires viewer attribute information related to the multiple viewers watching the video from information collected in advance by the attribute information collection unit 120 and stored in the attribute information storage unit 130. The comment information selection unit 170 selects comment information associated with a predetermined scene included in the video based on the viewer attribute information acquired in this way, and the comment output unit 200 outputs comments based on the selected comment information to the multiple user terminals 40 used by the multiple viewers watching the video. In this way, it is possible to output comments that are likely to attract the interest and attention of the viewers watching the video in real time.

[0094] Here, the common attribute information identification unit 150 may identify viewer attribute information common to the distributor attribute information of the video distributor from among the viewer attribute information acquired for the multiple viewers as the common attribute information. Also, the common attribute information identification unit 150 may identify attribute information common among the multiple viewers from among the viewer attribute information acquired for the multiple viewers as the common attribute information. For example, the common attribute information identification unit 150 may identify attribute information common among a predetermined number or more of viewers as the common attribute information, or may identify attribute information common among a predetermined ratio of viewers among the viewers watching the video as the common attribute information. The comment information selection unit 170 may select multiple comment information associated with a predetermined scene included in the video based on the common attribute information acquired in this way. In this way, it is possible to output comments that are likely to attract the interest and attention of multiple viewers watching the same video in real time.

[0095] (3) The multiple pieces of comment information associated with a specific scene included in a video may include comment information praising the distributor and comment information criticizing the distributor. The comment information selection unit 170 may preferentially select comment information praising the distributor or comment information criticizing the distributor, depending on the setting contents indicated in the setting information set by the distributor.

[0096] The comment information selection unit 170 may also infer the relationship between the viewer and the distributor based on the viewer attribute information and the distributor attribute information, and based on the inferred relationship, determine whether to preferentially select comment information praising the distributor or comment information criticizing the distributor. For example, if the viewer attribute information includes comment information indicating that the number of reactions to the distributor is equal to or greater than a threshold, it is inferred that the viewer is favorable to the distributor, so comment information praising the distributor may be preferentially selected. If the viewer attribute information and the distributor attribute information indicate that the viewer has lost to the distributor many times in a predetermined competitive game (the number of losses or the ratio of the number of losses to the number of wins is equal to or greater than a threshold), it is inferred that the viewer has a competitive spirit against the distributor, so comment information criticizing the distributor may be preferentially selected. In this way, it becomes possible to output comments that are more likely to attract the viewer's interest and attention in the video and the distributor more effectively.

[0097] (4) In the embodiment, an example has been described in which the comment information selection unit 170 selects at least one of the multiple pieces of comment information associated with each piece of event information E1 to E5 based on the viewer attribute information. Not limited to this, the comment information selection unit 170 may select at least one piece of comment information based on the viewer attribute information when it is determined that the multiple pieces of comment information associated with the event information are related to the viewer attribute information, and may not select comment information for the event information when it is determined that there is no comment information related to the viewer attribute information. In this case, the comment output unit 200 may output a comment together with a scene corresponding to the event information for which comment information has been selected, and may not output a comment together with a scene corresponding to the event information for which comment information has not been selected.

[0098] In the example shown in FIG. 7, the event information E3 is associated with comment information such as "A fan of a [professional baseball team], I will hunt for items at all costs.", "I'm desperate. I will hunt for even trash to become a [professional gamer].", and "I will devour it as if I were hunting for souvenirs from a trip to a [prefecture]." If the multiple pieces of viewer attribute information acquired by the attribute information acquisition unit 140 for a specific user do not include information on a professional baseball team, a professional gamer, or a travel destination, the comment information selection unit 170 determines that the comment information associated with the event information E3 does not have anything related to the viewer attribute information, and does not need to select comment information for the event information E3. In this case, the comment output unit 200 does not output a comment in a scene corresponding to the event information E3.

[0099] The viewer attribute information acquired by the attribute information acquisition unit 140 differs depending on the viewer. Therefore, scenes in which comments are output and scenes in which comments are not output differ depending on the viewer. In this way, by providing scenes in which comments are not output depending on the viewer, it is possible to prevent comments that do not attract the viewer's interest from being output, and to prevent the viewer's interest in the video from fading.

Claims

1. A video distribution device that distributes videos to viewers, an attribute information acquiring means for acquiring viewer attribute information which is attribute information indicating a tendency or track record of game play by a viewer who has accessed the video distribution device, or attribute information relating to a situation, behavior, or interest of the viewer which is inferred from content posted on an SNS service; a comment information selection means for selecting at least one of a plurality of pieces of comment information associated with a predetermined scene included in the video based on the viewer attribute information; a comment output means for outputting a comment based on the comment information selected by the comment information selection means together with the predetermined scene included in the video; A video distribution device including:

2. A video distribution device that distributes videos to viewers, an attribute information acquiring means for acquiring viewer attribute information indicating attributes of a viewer who has accessed the video distribution device; a comment information selection means for selecting at least one of a plurality of pieces of comment information associated with a predetermined scene included in the video based on the viewer attribute information; a comment output means for outputting a comment based on the comment information selected by the comment information selection means together with the predetermined scene included in the video; Including, A video distribution device characterized by at least one of the following features 1 to 4. Feature 1: the attribute information acquisition means acquires a plurality of pieces of viewer attribute information indicating attributes of the viewer; the video distribution device further includes a common attribute information identification means for identifying, among the plurality of viewer attribute information, viewer attribute information common to distributor attribute information indicating an attribute of a distributor of the video as common attribute information; The comment information selection means selects at least one of the plurality of comment information associated with the predetermined scene, based on the viewer attribute information identified as the common attribute information. Feature 2: The comment information includes a standard phrase, The comment output means outputs a comment including the fixed phrase included in the comment information and a statement of an attribute indicated by the viewer attribute information. Feature 3: The video distribution device further includes an interest level determination means for determining a level of interest of the viewer in an attribute indicated by the viewer attribute information, The comment information selection means determines at least one of the plurality of viewer attribute information corresponding to each of the plurality of comment information associated with the specified scene based on the plurality of interest levels determined for each of the plurality of viewer attribute information, and selects the comment information corresponding to the determined viewer attribute information. Feature 4: the comment information selection means, when it is determined that the plurality of pieces of comment information associated with the predetermined scene includes information related to the viewer attribute information, selects at least one of the plurality of pieces of comment information based on the viewer attribute information, and when it is determined that the plurality of pieces of comment information associated with the predetermined scene does not include information related to the viewer attribute information, does not select the comment information; When the at least one comment information is selected by the comment information selection means, the comment output means outputs a comment based on the selected comment information together with the specified scene, and when the comment information is not selected by the comment information selection means, the comment output means does not output a comment together with the specified scene.

3. A video distribution device according to claim 2, comprising the video distribution device including feature 1, The video distribution device further includes an interest level determination means for determining a level of interest of the viewer and a level of interest of the distributor in an attribute indicated by the viewer attribute information identified as the common attribute information, The comment information selection means determines at least one of the plurality of pieces of viewer attribute information corresponding to the plurality of pieces of comment information associated with the predetermined scene, based on the viewer's interest level and the distributor's interest level determined for each of the plurality of pieces of viewer attribute information, and selects the comment information corresponding to the determined viewer attribute information. Video distribution device.

4. A video distribution device according to claim 2, which includes the feature 3, or a video distribution device according to claim 3, the video distribution device further includes an emotion level determination means for determining an emotion level of the comment information based on the interest level determined for the viewer attribute information corresponding to the comment information; The comment output means outputs the comment based on the comment information in a manner corresponding to the emotional level. Video distribution device.

5. an attribute information acquiring step in which attribute information acquiring means acquires viewer attribute information which is attribute information indicating a tendency or achievement of game play by a viewer who has accessed the video distribution device, or attribute information relating to a situation, behavior, or interest of the viewer which is inferred from contents posted on an SNS service; a comment information selection step in which a comment information selection means selects at least one of a plurality of pieces of comment information associated with a predetermined scene included in the moving image based on the viewer attribute information; a comment output step in which a comment output means outputs a comment based on the comment information selected in the comment information selection step together with the predetermined scene included in the video; Video distribution methods including:

6. an attribute information acquisition step in which attribute information acquisition means acquires viewer attribute information indicating attributes of a viewer who has accessed the video distribution device; a comment information selection step in which a comment information selection means selects at least one of a plurality of pieces of comment information associated with a predetermined scene included in the moving image based on the viewer attribute information; a comment output step in which a comment output means outputs a comment based on the comment information selected in the comment information selection step together with the predetermined scene included in the video; Including, A video distribution method characterized by at least one of the following features 1 to 4. Feature 1: In the attribute information acquisition step, the attribute information acquisition means acquires a plurality of pieces of viewer attribute information indicating attributes of the viewer, The common attribute information identifying means further includes a common attribute information identifying step of identifying, as common attribute information, the viewer attribute information common to distributor attribute information indicating an attribute of a distributor of the video, among the plurality of viewer attribute information; In the comment information selection step, the comment information selection means selects at least one of the plurality of pieces of comment information associated with the predetermined scene, based on the viewer attribute information identified as the common attribute information. Feature 2: The comment information includes a standard phrase, In the comment output step, the comment output means outputs a comment including the fixed phrase included in the comment information and a statement of an attribute indicated by the viewer attribute information. Feature 3: The interest level determination means further includes an interest level determination step of determining a level of interest of the viewer in an attribute indicated by the viewer attribute information, In the comment information selection step, the comment information selection means determines at least one of the multiple pieces of viewer attribute information corresponding to the multiple pieces of comment information associated with the specified scene based on the multiple interest levels determined for each of the multiple pieces of viewer attribute information, and selects the comment information corresponding to the determined viewer attribute information. Feature 4: In the comment information selection step, when it is determined that the plurality of pieces of comment information associated with the predetermined scene includes a piece of comment information that is related to the viewer attribute information, the comment information selection means selects at least one of the plurality of pieces of comment information based on the viewer attribute information, and when it is determined that the plurality of pieces of comment information associated with the predetermined scene does not include a piece of comment information that is related to the viewer attribute information, the comment information selection means does not select the comment information, In the comment output step, if at least one comment information is selected in the comment information selection step, the comment output means outputs a comment based on the selected comment information together with the specified scene, and if the comment information is not selected in the comment information selection step, the comment output means does not output a comment together with the specified scene.

7. A video distribution program that causes a computer to function as a video distribution device that distributes videos to viewers, An attribute information acquiring means for acquiring viewer attribute information which is attribute information indicating the tendency or achievements of game play by a viewer who has accessed the computer, or attribute information relating to the situation, behavior, or interest of the viewer which is inferred from the contents of posts on an SNS service; a comment information selection means for selecting at least one of a plurality of pieces of comment information associated with a predetermined scene included in the video based on the viewer attribute information; and a comment output means for outputting a comment based on the comment information selected by the comment information selection means together with the predetermined scene included in the video; A video distribution program that causes the computer to function as a

8. A video distribution program that causes a computer to function as a video distribution device that distributes videos to viewers, an attribute information acquiring means for acquiring viewer attribute information indicating attributes of a viewer who has accessed the computer; a comment information selection means for selecting at least one of a plurality of pieces of comment information associated with a predetermined scene included in the video based on the viewer attribute information; and a comment output means for outputting a comment based on the comment information selected by the comment information selection means together with the predetermined scene included in the video; and causing the computer to function as A video distribution program characterized by at least one of the following features 1 to 4. Feature 1: the attribute information acquisition means acquires a plurality of pieces of viewer attribute information indicating attributes of the viewer; and further causing the computer to function as a common attribute information identification means for identifying, among the plurality of pieces of viewer attribute information, the viewer attribute information that is common to distributor attribute information indicating an attribute of a distributor of the video as common attribute information; The comment information selection means selects at least one of the plurality of comment information associated with the predetermined scene, based on the viewer attribute information identified as the common attribute information. Feature 2: The comment information includes a standard phrase, The comment output means outputs a comment including the fixed phrase included in the comment information and a statement of an attribute indicated by the viewer attribute information. Feature 3: causing the computer to further function as an interest level determination unit that determines a level of interest of the viewer in an attribute indicated by the viewer attribute information; The comment information selection means determines at least one of the plurality of viewer attribute information corresponding to each of the plurality of comment information associated with the specified scene based on the plurality of interest levels determined for each of the plurality of viewer attribute information, and selects the comment information corresponding to the determined viewer attribute information. Feature 4: the comment information selection means, when it is determined that the plurality of pieces of comment information associated with the predetermined scene includes information related to the viewer attribute information, selects at least one of the plurality of pieces of comment information based on the viewer attribute information, and when it is determined that the plurality of pieces of comment information associated with the predetermined scene does not include information related to the viewer attribute information, does not select the comment information; When at least one comment information is selected by the comment information selection means, the comment output means outputs a comment based on the selected comment information together with the specified scene, and when the comment information is not selected by the comment information selection means, the comment output means does not output a comment together with the specified scene.

Images