JP4512286B2 - Program sending system and program sending device used therefor - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to automated news program production, and more particularly, to a program transmission system in which production costs are reduced by using voice synthesized by a character-to-speech conversion system and a character image synthesized by CG technology, and the program transmission system. The present invention relates to a program transmission device.
[0002]
[Prior art]
Traditionally, news programs are pre-prepared by an editor who belongs to a broadcasting station, and news announcers who are seated at the news studio casters read out and shoot them with a TV camera to create video with audio. In addition, it was produced by keying the news material video prepared in advance as needed.
[0003]
[Problems to be solved by the invention]
However, announcers are required to have the ability to read news manuscripts at a reasonable speed and with clear pronunciation, and broadcasters need to bear a considerable amount of time and money to train announcers with such skills. As news content covers a wide range of genres, the announcer must appropriately state each news content in order to communicate the news with appropriate expressions such as facial expressions and intonation according to the news content. It is required to have a certain level of knowledge and education such as having to understand. In addition, because the audience rating of news programs depends on the viewer's preference for the announcer, the announcer must have a form and tone that can receive high likability from general viewers. From the broadcasting station and news program production side, it is not always easy to adopt and train announcers with the above qualities, and to establish a system that can always be placed in each news program. .
[0004]
In addition, the number of news programs is expected to increase in the future due to the increase in the number of channels and the diversification of news programs such as community-based programs. It is currently necessary to secure the necessary number of announcers with sufficient qualities. It is predicted that the production cost of each news program will need to be further reduced than it is today.
[0005]
Therefore, there is a need for a new news program production system that can reduce the cost of news program production and cope with a shortage of announcer personnel with sufficient qualities.
[0006]
The present invention has been made in view of the above-mentioned conventional problems or problems, and its purpose is to reduce the production cost of broadcast programs and to provide a new program that can cope with the shortage of performers such as announcers with sufficient qualities. It is to provide a transmission system and a program transmission apparatus used therefor.
[0007]
[Means for Solving the Problems]
The present invention has been made in view of the above problems, and as a program transmission system, the base voice, tone, pitch and speed of each broadcast program are designated based on the original data of the broadcast program, and each broadcast program is specified. Speech synthesis means for generating speech, character image generation means for generating a video of a performer of a program, and a mouth facial expression for adjusting the facial expression of the character image according to the sound output from the speech synthesis means Data and character image data; character image synthesis means for synthesizing a character image obtained by adding a modification based on mouth expression data to the character image data in accordance with the document data; and voice output from the voice synthesis means. Image / sound synthesis means for synthesizing the character images output from the character image synthesis means synchronously; Video wave storage means for converting the voice and character image output from the video / audio synthesis means into a television signal and broadcasting, and video data storage means for storing the collected video data, and video for switching the output of the video data Switching means, and the video image switching means selectively switches between the character image output from the character image synthesis means and the coverage video from the video data storage means. Further, according to the present invention, as the program transmission device, the character image generating means includes a plurality of character image data, and selects one character image according to the contents of the broadcast program .
[0008]
In the present invention, the manuscript data is the text type or other text data constituting the news content read out by the character serving as the announcer in the present invention. Control data for designating facial expression types such as emotions shown by images is added. Document data composing manuscript data is created by editing article data created from information obtained by a news reporter by contacting a news source into a form suitable for broadcasting as a news program. The control data is added according to the news content during editing.
[0009]
The voice synthesizing unit synthesizes a voice uttered by a person from document data constituting the manuscript data, and preferably has a function of adjusting a tone, a pitch, a speed, etc. according to control data for designating a voice type. ing.
[0010]
Also, the character image composition means modifies the character image data created based on a photographed image or animation of an actual person and the mouth periphery of the character image for each sound to be generated based on the CG technique. Mouth facial expression data for application, and a character image with a modification for each voice added to the character image or, if necessary, a modification according to the expression type such as emotions described above is output. It has a function.
[0011]
The voice output from the voice conversion means and the character image output from the character image synthesis means are synchronously synthesized by the image voice synthesis means, converted into a television signal, and broadcast to a general viewer.
[0012]
According to the present invention, the manuscript data is converted into voice by the voice synthesizing means and output, and at the same time, a character image that is CG-combined as an image is provided. A similar viewing experience can be given to viewers, which eliminates or reduces the need for broadcasters to always have announcers that require skills and qualities of a certain level or higher corresponding to each program. Thus, it is possible to achieve a reduction in costs such as personnel expenses and personnel education expenses.
[0013]
The document data in the present invention can also be created by editing the article data transmitted from the local personal computer carried by the reporter via the communication network by the in-house editing system.
[0014]
Article data is usually created by reporters interviewed in various parts of Japan where news content such as news sources exists, or overseas, but the reporter who conducted the coverage can carry it in the news source area. You can improve the efficiency of news article transmission by inputting article data collected on a local computer and transmitting it to a broadcasting station via a communication network. The intra-office editing system, which has the function of converting the body article data into a colloquial style suitable for the broadcast of the program, can further reduce costs such as personnel costs by creating the manuscript data from the article data mechanically Is possible.
[0015]
The program transmission system according to the present invention further includes video / audio storage means such as a VTR or CD for storing video data, and a video switcher for synthesizing a plurality of images. It is possible to provide a news program with a high sense of realism by synthesizing the news video from the VTR means with the voice and character image by keying synthesis by the video switcher and converting it into a television signal.
[0016]
Furthermore, the character image composition means of the present invention comprises a plurality of character images, and can be configured to select a character image corresponding to the news content and output the character image. It is possible to select a character image having an image, and to further improve the favorability received by the viewer.
[0017]
Furthermore, in the case of a program transmission system based on a bidirectional communication function, a character selection signal desired by the viewer is received from a receiver having a bidirectional communication function, and a character image according to the viewer's desire is used. Thus, it is possible to configure a broadcast program, which makes it possible to realize a broadcast program provided by a character image having an image adapted to the sensitivity of each individual viewer.
[0018]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing an example of a news broadcast system which is an embodiment of a program transmission system of the present invention.
[0019]
In FIG. 1, 1 is a local personal computer carried by a local reporter reporting at the news site, a word processor function that records and records the contents of articles interviewed by the local reporter, and communications such as mobile communications, public telephone lines, and the Internet. It has a modem function to connect to the network 2, and the article data created by the reporter on the local computer at the news source location etc. can be directly connected to the communication device such as the mobile phone 3 by the local personal computer 1. It is transmitted to the broadcasting station via the communication network 2.
[0020]
The broadcasting device in the broadcasting station includes a communication control unit 4 that controls the communication operation with the communication network 2, and an in-station editing system 5 that receives article data transmitted from the local personal computer 1 and edits a broadcast manuscript such as news. A CG announcer generating device 6 as a character image generating means for generating a video of a program participant such as an announcer by a computer graphic function, a background video generating device 7 for generating a background video corresponding to the news program, and the CG A video synthesis unit 8 that synthesizes the announcer video and the background video, and a voice synthesis unit 9 for generating the audio of the CG announcer are provided. The broadcast apparatus also includes a video / audio storage unit 10 for storing video data and audio data transmitted from the local personal computer 1, a telop generation unit 11 for generating a telop to be inserted into the video being broadcast, and a video synthesis unit. Video switcher 12 for switching video data and audio data from video / audio storage unit 10 or telop from telop generation unit 11 with respect to the video synthesized in 8, and video with respect to the audio synthesized in audio synthesis unit 9 An audio mixer 13 that inserts audio data from the audio storage unit 10 and a master system 14 that synchronizes the video output from the video switcher 12 and the audio output from the audio mixer 13 as radio waves for television broadcasting. ing. In FIG. 1, reference numeral 15 denotes a television camera, which performs an imaging operation for normal television broadcasting, and can synthesize the captured video in the video synthesis unit 8. Reference numeral 18 denotes a system control unit that controls the overall operation of the broadcasting apparatus in the broadcasting station.
[0021]
The in-office editing system 5 is an editor in charge within the office or a computer having an editing function, which selects article data sent from the local personal computer 1 as appropriate, classifies data, etc. The amount of text (number of characters) is edited into a format suitable for reading a news program. The intra-office editing system 5 preferably further includes base voice selection data for designating the base voice of the voice output from the voice synthesizer 9 according to the content of the news, voice for selecting and specifying the voice tone, pitch, and speed. A modification program as software, character selection data for specifying the type of character image output by the CG announcer generator 6, and a character for selecting and specifying facial expressions, gestures and gestures at the mouth of the character image Having modification programs as software, adding control data generated or processed by the respective modification programs to video data and audio data to create news manuscript data, It is sent to the speech synthesizer 9.
[0022]
The CG announcer generator 6 and the background generator 7 include an image database 6a that stores image data necessary for each operation. This image database 6a adjusts the expression of the mouth of the character image for the CG announcer generating device 6 in accordance with the image of the person as a model or the character image data created based on the animation and the output sound. Mouth facial expression data for storing image data for generating various background videos for the background generating device 7 is also stored. Here, the mouth expression data includes parameters for modifying character image data so that typical movements of the mouth when sounding each voice composing Japanese is reproduced on the character image by CG technology. It is made up of.
[0023]
The CG announcer generating device 6 reads necessary mouth expression data from the image database 5a according to the text information of the news manuscript data received from the in-house editing system 5 in addition to the basic character image data, and the mouth expression By adjusting and outputting the character image data based on the data, a character image that appears as if the character naturally pronounces the sound output from the speech synthesizer 9 is output.
[0024]
Further, the CG announcer generating device 6 is based on the character modification program received from the in-house editing system 5 so that the character is further modified so that the character performs more natural expressions, gestures, hand gestures, and the like. It can also be configured to output image data.
[0025]
Further, the image database 6a can include a plurality of character image data and mouth expression data corresponding to each character image data, and the CG announcer generating device 6 receives from the in-office editing system 5 or both In the case of a bidirectional reception system, it is also possible to use character image data selected based on character selection data received from a receiver having a bidirectional communication function, and to output the character image data obtained by modifying the character image data. It is.
[0026]
The voice synthesizer 9 is based on all the voices that make up the Japanese language by adding muddy sounds to the so-called 50 sounds of “Aiueo ...” for each of the base sounds sampled from a plurality of men and women. The voice database 9a is stored as a file for each voice. Then, based on the text information of the news manuscript data, the voice synthesizer 9 sequentially extracts necessary voices from the file in the voice database 9a in which the base voice designated by the base voice designation data is stored, and synthesizes a voice signal. By doing so, it has a function of converting news manuscript data into sound. At this time, by adjusting the tone, level, speed, etc. of the synthesized voice signal on the computer based on the voice modification program, a more natural and expressive news voice can be obtained.
[0027]
The video / audio storage unit 10 has a function and a role as an external auxiliary storage device that stores video data and audio data, and is configured by a storage medium such as a VTR 16 or a CD 17. In addition, a larger capacity storage medium such as a hard disk or a DVD can be used.
[0028]
An operation when broadcasting a news program in the broadcasting system having such a configuration will be described.
[0029]
First, article data created by a reporter on a local computer 1 at a news source site or the like is transmitted to a broadcasting station via the communication network 2 by the local computer 1 connected directly or to a communication device such as a mobile phone 3. Is done. In addition, an on-site video image taken by the reporter or cameraman is also transmitted to the broadcasting station via the communication network 2. In the broadcasting station, the received article data is taken into the intra-office editing system 5 under the control of the system control unit 18, while the video is stored in the VTR 16 or CD 17 of the video / audio storage unit 10.
[0030]
Thereafter, the program creator of the broadcast station edits the article data using the intra-station editing system 5 to create news manuscript data. At this time, the news manuscript data and the video (captured video) stored in the video / audio storage unit 10 and corresponding to (or related to) the news manuscript are linked. Further, in this editing work, a decision is made as to whether a person who reads a news manuscript in a news program is an actual announcer or a CG announcer, and when selecting a CG announcer, which CG announcer is used is selected. When a CG announcer is used, character image and audio data relating to the CG announcer is also incorporated into the news manuscript data.
[0031]
In the case of using a CG announcer, when broadcasting of a news program starts, various control signals for performing broadcasting are output from the system control unit 18, and news manuscript data is transmitted from the in-house editing system 5. The CG announcer data is sent to the CG announcer generator 6, and the voice data is sent to the voice synthesizer 9. In addition, a control signal for generating a background video is sent to the background generation device 7.
[0032]
The CG announcer generating device 6 includes video or character image data of a person (CG announcer) designated from the image database 6a, and facial expression data for adjusting the facial expression of the CG announcer's mouth according to the output voice. Are read out and subjected to predetermined processing to output video data. Here, the predetermined process is to reproduce typical movements of the mouth when the sounds constituting Japanese are pronounced on the character image by CG technology, as if the CG announcer is reading a manuscript or news This is a process of modifying the character image data so as to explain the above. Further, as a part of the processing, the CG announcer generating device 6 is configured so that the character performs more natural facial expressions, gestures, hand gestures, and the like based on the character modification program received from the local editing system 5. Furthermore, character image data with further modifications can be output. The background generation device 7 generates a background video suitable for the background of the CG announcer as a video of the news program.
[0033]
The voice synthesizer 9 reads the designated voice data from the voice database 9a, performs a predetermined voice synthesis process to generate a CG announcer voice, and outputs the voice data. The voice synthesizer 9 converts the news manuscript data into voice based on the text information of the news manuscript data in the voice synthesizing process. At this time, by adjusting the tone, level, speed, etc. of the synthesized voice signal on the computer based on the voice modification program, a more natural and expressive news voice is obtained. The output from the voice synthesizer 9 is sent to the voice mixer 13 where it is mixed with other voices, such as sound effects and background music, and output as overall voice data.
[0034]
The output from the video composition unit 8 is sent to the video switcher 12. Whether the video switcher 12 should send a CG announcer video as a video to be broadcast, a news scene image from the video / audio storage unit 10, or a telop video from the telop generator 11 Is determined by a control signal from the system control unit, and an output video signal is selected.
[0035]
After performing these processing operations, the output signals from the speech synthesizer 9, the CG announcer generator 6 and the background generator 7 are sent to the master system 14, and transmitted from this release master system 14 as radio waves for television broadcasting. The
[0036]
As a result, when it is difficult to always deploy in the station or when it is desired to reduce the cost of program production, a virtual created by the speech synthesizer 9 and the CG announcer generator 6 is used without using an actual announcer. A broadcast system for a news program can be constructed in which a news announcer is displayed as a video and the virtual announcer delivers news with a natural tone and motion expression that is not different from a real announcer.
[0037]
Since the present program transmission system includes a VTR 16 and a CD 17 for storing news video data photographed in advance and a video switcher 12 for switching a plurality of images, the video signal synthesized by the video synthesis unit 8 is provided. The news video stored in the VTR 16 or the like is appropriately switched and output using the video switcher 7 so that the actual news video can be broadcasted, and a program transmission system with a higher presence can be provided.
[0038]
Further, studio image data obtained by reproducing a set in the studio is further stored in the above-described image database 6a, and the video composition unit 8 uses the character image data obtained by superimposing the studio image data as the background of the character image. It is also possible to output the information, which makes it possible to convey the atmosphere unique to the news program to the viewer.
[0039]
In this embodiment, when the CG announcer is not used, a news program is displayed while a real announcer is introduced in front of the TV camera 15 and the announcer reads a manuscript or explains the news. Carry out broadcasting.
[0040]
In the above embodiment, the news program is taken up as a broadcast program. However, the present invention can be applied to various broadcast programs such as a news commentary program, a document program, a conversational program, or a political broadcast. In particular, the program transmission system according to the present invention has the advantage of reducing costs by reducing the number of people who appear in the program, but is not limited thereto, and it is not possible to appear for any reason if you want to appear (or want to) appear in a TV program. CG performers corresponding to the above CG announcers can be generated and broadcast about people, and problems and problems in a busy society like the present can be solved.
[0041]
In addition, a plurality of CG announcers are provided in the CG announcer generating device 6 and the voice in the speech synthesizer 7 can generate a plurality of types of audio data, and both the broadcasting station and the receiving device on the viewer side are provided. It is also possible to enable communication to the viewer so that the viewer can select the CG announcer and voice according to his / her preference.
[0042]
【The invention's effect】
As described above, according to the present invention, all or part of a program such as news using actual performers that are difficult to constantly deploy in the station is created by the speech synthesis unit and the character image synthesis means. It is possible to create a program transmission system that delivers virtual human images with natural motion expressions that do not differ from actual performers, reducing broadcast program production costs and making viewers more likable Broadcasting using characters and voice that can be given is possible.
[Brief description of the drawings]
FIG. 1 is an explanatory diagram showing an embodiment of the present invention.
[Explanation of symbols]
1 Local PC 2 Communication network 3 Mobile phone (communication equipment)
4 Communication Control Unit 5 Intra-Station Editing System 6 CG Announcer Generation Device 6a Audio Database 7 Background Generation Device 8 Video Synthesis Unit 9 Audio Synthesis Unit 9a Audio Database 10 Video / Audio Storage Unit 11 Telop Generation Device 12 Video Switcher 13 Audio Mixer 14 Master System 15 TV camera 16 VTR
17 CD
18 System controller

Claims (2)

Voice synthesis means for generating voice for each broadcast program by designating base voice, tone, pitch and speed for each broadcast program based on the original data of the broadcast program ;
Character image generating means for generating video of performers of the program;
Mouth expression data and character image data for adjusting the expression of the mouth of the character image according to the sound output from the speech synthesis means,
Character image synthesis means for synthesizing a character image obtained by adding a modification based on mouth expression data to the character image data according to the original data;
Image / sound synthesis means for synthesizing the voice output from the voice synthesis means and the character image output from the character image synthesis means in synchronism;
Broadcast radio wave transmission means for converting the sound and character image output from the image sound synthesis means into a television signal and broadcasting it,
Video data storage means for storing the video data collected, and video switching means for switching the output of the video data, and the character image output from the character image synthesis means and the coverage video from the video data storage means A program transmission system, wherein selection switching is performed by a switching means. 2. The program transmission system according to claim 1, wherein the character image generation means includes a plurality of character image data, and selects one character image according to the contents of the broadcast program .

Images