JPH05176232A - Title superimposing device - Google Patents

Abstract

(57) [Abstract] [Purpose] Converts an audio signal into character data and superimposes it on a video signal. [Structure] A voice signal 2 from a television or the like is converted into a character code by voice recognition, processed by a sentence generation unit 4 and a character generation unit 5, and temporarily stored in a video memory 6 as character pattern data. The subtitle synthesizer 8 superimposes the character pattern data on a predetermined position of the video signal based on the synchronization signal detected from the video signal.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a caption superimposing device,
The present invention relates to a device for converting an audio signal of a video / audio signal of a television or the like into character data and superimposing it on a video signal.

[0002]

2. Description of the Related Art Conventionally, in television broadcasting and the like, audio information is added together with a video signal by a video signal by inserting subtitles, a broadcast in which a sign language screen is inserted in a corner of a main screen,
Alternatively, there is a character information service or the like that complements a table program in a character broadcast. However, these were limited to some programs, and most of them consisted only of video and audio information. Deaf people are strongly required to insert subtitles when playing all TV programs and VTRs, and also use subtitles to help learn languages in bilingual broadcasting by voice multiplexing. Great value. Therefore, there has been a demand for a device that converts audio information from video / audio information into text information and superimposes it on video information.

[0003]

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned problems. An audio signal is taken out from a video / audio signal, a voice recognition process is performed, and these are converted into character data into a video signal. A subtitle superimposing device for superimposing is provided.

[0004]

In order to solve the above-mentioned problems, the present invention takes out an audio signal from a video / audio signal of a television or the like and converts the audio signal into a character code by voice recognition, and A character generator that outputs corresponding character data based on the same character code, a video memory that temporarily stores the character data, and a caption synthesizer that superimposes the character data from the video memory on a predetermined position of the video signal. The subtitle superimposing device is provided.

[0005]

With the above configuration, in the caption superimposing device according to the present invention, an audio signal is extracted from the input video / audio signal, this audio signal is converted into a character code by voice recognition, and this character code is converted into a character code. It is converted into a pattern and superimposed on a predetermined position of the video signal.

[0006]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT An embodiment of a caption superimposing device according to the present invention will be described below in detail with reference to the drawings. In the figure, 1 is a video signal and 2 is an audio signal, which is input from a television, a VTR or the like. A voice recognition unit 3 converts an input voice signal into a character code. Reference numeral 4 denotes a sentence generation unit that outputs a character code converted into an appropriate sentence based on the character code from the voice recognition unit 3. A character generator 5 converts the character code from the sentence generator 4 into character pattern data. A video memory 6 temporarily stores character pattern data from the character generator. Reference numeral 7 denotes a sync signal detector which separates the sync signal from the video signal and outputs it. Reference numeral 8 denotes a caption synthesizing unit that synthesizes character data and a video signal from the video memory 6. A control unit 9 controls each of the above units. A video output signal 10 is superimposed on the caption signal and is output to a display device such as a television together with the audio output signal 11.

Next, the operation of the subtitle superimposing device according to the present invention will be described. For the audio signal 2, the main audio signal is selected in the case of main / sub audio in audio multiplex broadcasting or the like, and is converted to monaural in the case of stereo, and the audio signal for subtitle display is input to the voice recognition unit 3. The voice recognition unit 3 extracts a characteristic parameter from an input voice signal from which environmental sounds and the like have been removed via a voice filter, compares it with a standard pattern stored, and performs recognition processing in syllable units, for example, and converts it into a character code. . The sentence generation unit 4 stores linguistic information such as syntax, meaning, and context. Based on this linguistic information, the input character code is controlled to be converted into a sentence composed of appropriate words, idioms, etc., and the corresponding character code is converted. Output sequentially. In the character generation unit 5, character pattern data corresponding to each character code is stored, read out based on the character code from the sentence generation unit 4, and stored in the video memory 6. The subtitle synthesizer 8 uses the horizontal / vertical sync signal from the sync signal detector 7 to detect the video memory 6
The character pattern data stored in is read out, superimposed on a predetermined position of the video signal, and the video signal for caption insertion is output.

FIG. 2 is a block diagram of a main part of another embodiment of the caption superimposing apparatus according to the present invention, in which a delay means 12 for video signals and audio signals is provided. The delay means 12 is, for example, a BBD.
Using a delay element such as (Bucket Brigade Device), the video signal delayed for a predetermined time (audio signal processing time) is output to the caption synthesis unit 8, and the audio signal is output as the audio output signal 11 as it is. Or if the required delay time is large,
A writing head and a reading head are provided. First, the video signal and the audio signals 1 and 2 are written on the endless recording medium by the writing head, and after a predetermined time has passed, the video signal is read by the reading head and output to the subtitle synthesizer 8. Is output as it is as the audio output signal 11. By doing so, it is possible to superimpose a subtitle in synchronization with the video signal.

[0009]

As described above, in the caption superimposing device according to the present invention, the audio signal is extracted from the input video / audio signal for voice recognition, and the converted character data is superimposed on the video signal and output. Therefore, if this signal is connected to a video / audio input terminal of a television or the like, an image with subtitles can be displayed. Therefore, it is possible to deal with all information such as video and audio of a TV, VTR, etc., and thus it is possible to provide the hearing-impaired person with image information to which character information is added. In addition, it can be used for language learning, etc., such as listening to a foreign language by voice and checking it with subtitles in a multiplex audio broadcast in two languages.

[Brief description of drawings]

FIG. 1 is a block diagram of a main part of an embodiment of a caption superimposing device according to the present invention.

FIG. 2 is a block diagram of a main part of another embodiment of the caption superimposing device according to the present invention.

[Explanation of symbols]

 1 Video signal 2 Audio signal 3 Speech recognition unit 4 Text generation unit 5 Character generation unit 6 Video memory 7 Subtitle synthesis unit 8 Sync signal detection unit 9 Control unit 10 Video output signal 11 Audio output signal 12 Delay means

Claims (2)

[Claims] 1. A voice recognition unit that extracts a voice signal from a video / voice signal of a television or the like, converts the voice signal into a character code by voice recognition, and a character generation unit that outputs corresponding character data based on the same character code. And a video memory for temporarily storing character data, and a caption synthesizing unit for superimposing character data from the video memory at a predetermined position of the video signal. 2. The caption superimposing apparatus according to claim 1, wherein the video signal is input to the caption synthesizing unit via a delay unit so as to be synchronized with character data for superimposition.