JPWO2014196457A1 - Content playback device with subtitles - Google Patents

Abstract

The content playback apparatus with subtitle performs translation based on the subtitle analysis unit that extracts the subtitle character string and the auxiliary information for translation included in the subtitle content, and the extracted subtitle character string and the auxiliary information for translation. A translation execution unit; and a translation character string output unit that outputs a translation result generated by the translation execution unit as a character string.

Description

  The present invention extracts a subtitle character string from content including a subtitle stream, obtains a result obtained by translating the subtitle character string into a language different from the language, and reproduces content with subtitles for displaying the result Relates to the device.

  It has been a long time since digital broadcasting that compresses video and audio as digital data and multiplexes and broadcasts has been put into practical use. In digital broadcasting, it is possible to display the content currently being broadcast as captions in consideration of the hearing impaired and elderly people, but it may be troublesome if captions are always superimposed on the broadcast. For this reason, it is necessary to be able to perform ON / OFF control of subtitle display. Therefore, the subtitles are not broadcast in a form superimposed on the video, but are multiplexed as a subtitle stream independently of the video stream, and display ON / OFF is possible.

  By the way, with the enhancement of the performance of machine translation engines, there is a demand for translating subtitles included in the above-described digital broadcasting into an arbitrary language and displaying them. However, subtitles are displayed at a frequency of about once every few seconds, but the display may change in the middle of sentences and words due to limitations on the display area. If machine translation is performed for each displayed sentence, it will not be translated correctly. And the phenomenon will occur.

  Thus, paragraph 0059 of Patent Document 1 discloses a technique of “determining the end of a sentence with a. (Period) or the like and sending a plurality of items to the machine translation unit as one sentence”.

Japanese Patent Publication “JP 2009-016910 A”

  However, a period or a punctuation mark is a sentence delimiter, but another sentence may be displayed from the next screen without a punctuation mark. In addition, when using commas or punctuation marks, there are cases where sentences are separated correctly or not. That is, it is difficult to detect sentence breaks for correct translation from information acquired as character strings.

  For example, FIG. 7 shows a scene introducing a family in a TV program. First, as shown in (A), an image of the eldest son's up and a subtitle “This is the eldest son” are displayed, and a few seconds later. As shown in (B), the second son ’s up video and the subtitle “This is the second son” are displayed. A few seconds later, the third son ’s up video and the subtitle “This is the third son” appear. This is a scene in which, after a few seconds, three videos and a subtitle “This is all children” are displayed as shown in (D).

  In this case, it is desirable to translate the character string up to the punctuation mark and output the translation result without waiting for the punctuation mark so that the three people's video and the relationship have a one-to-one correspondence.

  Also, in a different scene, as shown in (E), the video of the eldest son ’s up and the subtitle “The eldest son is the third son” are displayed. And there is a scene where the subtitles say “I am smart”.

  In this case, in order to clarify the subject and the object, it is desirable to translate the character string up to the punctuation point instead of the punctuation point and output the translation result.

  In addition, proper nouns such as names and places cannot be matched in the program unless they are always translated into the same language, which hinders understanding, but no solution to this problem has been presented.

  For example, Fig. 8 shows a scene of a drama with a character named Hayashi. In (G), the answer to the question "Who did that?" This is the scene to be displayed. In this case, when the character string “It is a forest” in (H) is translated, generally, the word “forest” is translated as a forest forest.

  In addition, when there is a scene in which a subtitle “My name is Yoshio Hayashi” is displayed as in (I), it is often translated as a proper name noun Hayashi. Then, when translated as a forest of forests and when translated as a forest of personal name nouns, the translated words differ, and it cannot be understood that they indicate the same thing.

  In order to solve the above-described problem, the first means of the present invention includes a subtitle analysis unit that extracts subtitle character strings and translation auxiliary information included in subtitle content, the extracted subtitle character strings, and translation subtitles. A translation execution unit that performs translation based on the auxiliary information and a translation character string output unit that outputs a translation result generated by the translation execution unit as a character string are provided.

  According to the present invention, information indicating a character string delimiter that a desired translation result is obtained by translating with a character string acquired so far, included in a subtitle stream, or a proper noun By translating using the start and end information, the translation can be performed in correct sentence units, and proper nouns are always translated into the same words, thereby improving the accuracy of machine translation.

It is a block diagram which shows schematic structure of an example of the content reproduction apparatus with a caption concerning the 1st Embodiment of this invention. It is a figure which shows the data structure of the caption stream prescribed | regulated by STD-B24. This is a code system of 8-unit character code defined in STD-B24. It is the table | surface which described some code formats of the extended control code defined by STD-B24. It is a list of extended control codes defined in STD-B24. It is a code sequence when auxiliary information for translation is defined as an extended control code. It is a figure explaining the scene which introduces a certain family in a television program. It is a figure explaining the scene of a drama with a television program. It is a figure explaining the code sequence 1 of the subtitle stream corresponding to the scene which introduces a certain family in a television program. It is a figure explaining the code sequence | part 2 of the subtitle stream corresponding to the scene which introduces a certain family in a television program. It is a figure explaining the code sequence 1 of a subtitle stream corresponding to the scene of a drama with a TV program. It is a figure explaining the code sequence 2 of a subtitle stream corresponding to the scene of a drama with a television program. It is a flowchart which shows the flow of a process of a caption stream in the content reproduction apparatus with a caption of this invention. It is the table | surface which showed the transition of the data by the process using the information of personal name proper noun start information, and the information of personal name proper noun end. It is a block diagram which shows schematic structure of an example of the content reproduction apparatus with a caption concerning the 2nd Embodiment of this invention. It is a block diagram which shows schematic structure of an example of the content reproduction apparatus with a caption concerning the 3rd Embodiment of this invention. It is a block diagram which shows schematic structure of an example of the content reproduction apparatus with a caption concerning the 4th Embodiment of this invention.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
[Embodiment 1]
FIG. 1 is a block diagram showing a schematic configuration of an example of a content reproduction apparatus with subtitles according to the first embodiment of the present invention.

  As shown in the figure, the content reproduction apparatus with subtitle 1001 includes a content reception unit 101, a stream separation unit 102, a video stream decoding unit 103, an audio stream decoding unit 104, a subtitle stream decoding unit 105, a subtitle character string acquisition unit 106, A translation auxiliary information acquisition unit 107, a translation destination language determination unit 108, a translation execution unit 109, a video output unit 110, an audio output unit 111, and a translated character string output unit 112 are provided.

  The content receiving unit 101 is a part that connects to a digital broadcast tuner (not shown), a communication network, or a recording medium that stores content, and acquires content including a subtitle stream.

  The stream separation unit 102 is a part that separates the content acquired by the content reception unit 101 into a video stream, an audio stream, and a subtitle stream.

  The video stream decoding unit 103 is a part that generates display video data by decoding the video stream separated by the stream separation unit 102.

  The audio stream decoding unit 104 is a part that generates output audio data by decoding the audio stream separated by the stream separation unit 102.

  The subtitle stream decoding unit 105 is a part that decodes the subtitle stream separated by the stream separation unit 102 to generate a subtitle character code string and a control code string.

  The subtitle character string acquisition unit 106 is a part that analyzes the subtitle character code sequence generated by the subtitle stream decoding unit 105 and acquires the subtitle character string.

  The translation auxiliary information acquisition unit 107 is a part that analyzes the subtitle control code sequence generated by the subtitle stream decoding unit 105 and acquires auxiliary information for translation.

  Here, although the subtitle character string acquisition unit 106 and the translation auxiliary information acquisition unit 107 are described as separate and independent components, the subtitle analysis unit having a function of combining both of them is used to reproduce the content with the subtitle. It is good also as a structure which implement | achieves an apparatus.

  The translation destination language determination unit 108 is a part that determines a translation destination language in response to an instruction from a user or the like (not shown).

  The translation execution unit 109 uses the translation auxiliary information acquired by the translation auxiliary information acquisition unit 107 to translate the subtitle character string acquired by the subtitle character string acquisition unit 106 by the translation destination language determination unit 108. This is the part that translates to

  The video output unit 110 is a part that outputs the display video data generated by the video stream decoding unit 103.

  The audio output unit 111 is a part that outputs the output audio data generated by the audio stream decoding unit 104.

  The translation character string output unit 112 is a part that outputs the translation result generated by the translation execution unit 109 as a character string. The character string of the translation result may be output in the form of being overlaid on the display video data output by the video output unit 110, or may be output to a display part different from the part output by the video output part 110. For example, a 16:10 display is used to display the video output by the video output unit 110 in the upper 16: 9 area, and the characters output by the translated character string output unit 112 in the lower 16: 1 area. A column may be displayed.

  As will be described later, the subtitle stream includes display time information, but there are cases where the subtitle stream is delayed and cannot be displayed at the designated display time. In that case, it may be displayed when the translation is completed. If the translation takes too much time, the translation of the sentence may be skipped and the next sentence may be translated.

  In order to play back content in the content playback apparatus with subtitles, a subtitle stream including a subtitle character string and auxiliary information for translation is generated in the content distribution apparatus, and the generated subtitle stream, video stream, and audio stream are generated. May be created and distributed to a content reproduction apparatus with captions.

  In order for the translation auxiliary information acquisition unit 107 to acquire the translation auxiliary information from the subtitle control code string, the information needs to be multiplexed in the subtitle stream in accordance with a predetermined rule. At that time, it is necessary to multiplex the existing rules so that the information can be skipped by an existing terminal that does not know the rules for auxiliary information for translation.

  Here, a method for adding ARIB standard STD-B24, which is a standard for Japanese digital broadcasting subtitle stream, to add auxiliary information for translation will be described. You may define additional auxiliary information for translation.

  FIG. 2 is a diagram illustrating a data structure of a subtitle stream defined by the existing STD-B24.

  The subtitle stream is stored in a unit called PES (Packetized Elementary Stream) for each unit displayed on the same screen, and the PES is composed of a PES header 11 and PES data 12.

  The PES header 11 includes display time information so that it can be interpreted at what time the subtitles in the subsequent PES data should be displayed.

  The PES data 12 includes a data group header 13, data group data 14, and CRC 15. The data group data is also called caption text data.

  The data group header 13 defines the type of subtitle text data that follows. Whether it is subtitle management data that specifies the display method, the number of languages, etc., or what language subtitle text data is included. Interpretation is possible.

  The caption text data 14 includes a caption text data header 16 and a data unit 17.

  The caption text data header 16 includes language information of the subsequent data unit.

  The data unit 17 includes a data unit header 18 and data unit data 19.

  The data unit data 19 includes a control code string 20a and a character code string 21a.

  The control code string 20a includes a plurality of pieces of information for controlling display of a character string such as the display position, color, and character decoration of the subsequent character code string 21a.

  The character code string 21a includes information representing a plurality of characters to be displayed as subtitles by a character code of an 8-unit character code defined by STD-B24.

  The control code string 20b is inserted in the middle of the data unit data 19 so that the subtitles displayed on one screen can be changed in the middle, and the character of the character code string 21b subsequent thereto. The structure can include information for controlling the display of the columns.

  FIG. 3 shows a code system of 8-unit character code defined by STD-B24. 1-byte data is divided into upper 4 bits and lower 4 bits, and is represented in a matrix.

  A region from 0x00 to 0x1F is called a C0 control code region, and a region from 0x80 to 0x9F is a region called a C1 control code region. Values in these two regions are included in the control code strings 20a and 20b in FIG. This is the start code of each control code.

  The control code consists of a parameter of the number of bytes determined for each control code after this start code, and constitutes one control code. Since this parameter does not use the values of the C0 control code area and the C1 control code area, even if a control code is newly defined, the existing device can skip to the start code of the next control code.

  0x21 to 0x7E is a region called a GL code region, and 0xA1 to 0xFE is a region called a GR code region. By using control codes, each of these two regions has a Chinese character set, an alphanumeric set, a hiragana set, A character string included in the character code strings 21a and 21b in FIG. 2 is shown by appropriately assigning a katakana set, an external character code set, an additional symbol set, and the like.

  These sets are divided into a set of 1-byte codes and a set of 2-byte codes. Even in the case of 2-byte codes, the same GL code area or GR code area is used for both the first byte and the second byte. It has become.

  CSI, which is a value of 0x9E in the C1 control code area, is positioned as an extended control code. Some extended control codes have already been defined, and FIG. 4 shows the format of some of the codes.

  As can be seen by looking at the SDF part of the fifth column from the CCC of the second column, each code sequence starts with 0x9E which is the start code of CSI, and is a parameter between 0x30 to 0x39 and an intermediate between 0x3B or 0x20 An arbitrary number of characters follow, and finally, a control code is composed of terminal characters different for each column. Some of them consist only of CSI start codes and end characters, as in the sixth and seventh columns. Also, like the SCS in the last column, a CSI start code and end character are defined for skipping, and the contents are not defined.

  FIG. 5 is a list of currently defined extended control codes, showing the terminal character code, intermediate character code, and possible values of parameters for each code.

  In both cases, the end character code is 0x40 to 0x6F, the intermediate character code is 0x20 or 0x3B, and the parameters are 0x30 to 0x39. Even when a new extended control code is defined, this rule is maintained, and 0x40 to If a terminal character code that has not yet been assigned in 0x6F is used, the existing device can skip to the start code of the next control code.

  Up to this point, the definition of the subtitle stream in the existing standard is described. Hereinafter, a method for additionally defining auxiliary information for translation in the existing standard will be described.

  FIG. 6 shows a code sequence when the auxiliary translation information is defined as an extended control code. 0x6D is assigned as the terminal character code, 0x30 is used as information for delimiter for translation execution, 0x31 is used as information on personal name proper noun start, 0x32 is used as information on end of personal name proper noun, and 0x33 is used as information on start of place name proper noun , 0x34 is assigned to the place name proper noun end information, 0x35 is assigned to other proper noun start information, and 0x36 is assigned to other proper noun end information.

  Translation execution delimiter information is information indicating that the character string delimiter is such that a desired translation result can be obtained by translating with a character string acquired by analyzing the character code string acquired so far. is there. In general, it is considered that the translation delimiter information is inserted at the end of a sentence or at the end of the screen. Therefore, information that is at the end of the sentence or at the end of the screen but is not a delimiter, that is, information for deferring translation execution, may be defined to substitute for the delimiter information for translation execution.

  The name proper noun start information is information indicating that the name proper noun starts from the character string of the next character code string, and the person name proper noun end information is the character code string string already received. This is information indicating that the proper noun of the person name has ended. That is, the character code string of the part sandwiched between the personal name proper noun start information and the personal name proper noun end information is the proper noun of the personal name.

  The place name proper noun start information and the place name proper noun end information are information indicating the start and end positions of the place name proper nouns, as in the above-described personal name proper noun start information and personal name proper noun end information.

  In addition, other proper noun start information and other individual noun end information are similar to the above-mentioned personal name proper noun start information and personal name proper noun end information. This is information indicating the position.

  In this case, the proper nouns are classified into three types: personal names, place names, and other types, and different parameter values are assigned. However, these may be set as one value without being classified, or may be classified more finely. . In addition, the start parameter may be unified into one value, and the end parameter may be divided into three to enable discrimination of three types of proper nouns. Conversely, the end parameter is set to one. Thus, it may be possible to discriminate three types of proper nouns with the start parameter. The merit of classifying proper nouns will be described later.

  Further, the translation execution delimiter information and the proper noun start / end information are distinguished by parameters using the same control code, but they may be defined as different control codes. These auxiliary information for translation are defined as extended control codes in the framework of CSI. However, they are defined by using undefined start codes that are shaded in the coding system of FIG. Also good.

  An example of the encoded stream of the caption stream extended and defined in this way will be described.

  FIG. 9 is a code stream of a subtitle stream corresponding to (A) to (D) of a scene introducing a certain family in FIG.

  The scene where the subtitle “This is the eldest son” in FIG. 7A corresponds to the code string of the data unit data 1 in FIG. 9 and is after the control code string such as the display position and the display color. , "This is the eldest son" followed by a character code string, followed by a translation execution delimiter control code.

  The scene where the subtitle “This is the second son” in FIG. 7B corresponds to the code string of the data unit data 2 in FIG. 9 and the control code string such as the display position and the display color. Is followed by a character code string “This is the second son,” followed by a translation execution delimiter control code.

  Then, the scene where the subtitle “This is the third son” in FIG. 7C corresponds to the code string of the data unit data 3 in FIG. 9 and the control code string such as the display position and the display color. Is followed by a character code string “This is the third son,” followed by a translation execution delimiter control code.

  Then, the scene in which the subtitle “This is all of the children.” In FIG. 7D corresponds to the code string of the data unit data 4 in FIG. 9 and controls the display position, display color, etc. After the code string, a character code string “This is all children.” Is followed by a control code for translation execution.

  As described above, when the translation control delimiter control signal is extracted, translation is performed using the character string obtained so far, so that the video of the three people and the sentence of the related translation result are output in a one-to-one correspondence. Become so.

  FIG. 10 is a code stream of a subtitle stream corresponding to (E) to (F) of a scene introducing a family shown in FIG.

  The scene where the subtitle “the eldest son is the third son” in FIG. 7E corresponds to the code string of the data unit data 5 in FIG. 10 and the control code string such as the display position and the display color. Is followed by a character code string “the eldest son is the third son”, and here, there is no control code for translation execution delimiter.

  The scene in which the subtitle “It is said to be clever” in FIG. 7F corresponds to the code string of the data unit data 6 in FIG. 10, and the control code string such as the display position and the display color. Is followed by a character code string saying "I am clever", followed by a translation execution delimiter control code.

  As described above, when the translation execution delimiter control signal is extracted, the translation is performed with the character string acquired so far, so that the translation result is output with the subject and the object being clarified.

  FIG. 11 is a code stream of a subtitle stream corresponding to (H) of one scene of a drama in which the character “Hayashi” in FIG.

  The scene where the subtitle “It is a forest” in (H) of FIG. 8 corresponds to the code string of the data unit data 7 of FIG. 11, after the control code string such as the display position and the display color. , Followed by the character code string “It is”, followed by the control code for the personal name proper noun start, followed by the character code string “Hayashi”, followed by the control code for the personal name proper noun end, “ Followed by a control code for a translation execution delimiter.

  As described above, since the control signal for starting the name proper noun and the control signal for ending the name proper noun are added, the proper noun is not translated as a common noun.

  FIG. 12 is a code stream of a subtitle stream corresponding to (I) of another scene of the drama with the character “Hayashi” in FIG.

  The scene where the subtitle “My name is Yoshio Hayashi” in (I) of FIG. 8 corresponds to the code string of the data unit data 8 of FIG. 12 and controls the display position, display color, etc. After the code string, the character code string “My name is” follows, followed by the control code for starting the name proper noun, and then the character code string “Yoshio Hayashi”, for controlling the end of the name proper noun. The code continues, followed by the character code string “is”, and finally the translation control delimiter control code.

  As described above, since the control signal for starting the name proper noun and the control signal for ending the name proper noun are added, the proper noun is not translated as a common noun.

  FIG. 13 is a flowchart showing the flow of processing of a caption stream in the content reproduction apparatus with caption according to the present invention. Here, the description is focused on the processing after the extraction of the data unit data 19 in FIG.

  First, in S01, data unit data in the caption PES data is extracted and processing is started.

  Next, in S02, it is confirmed whether or not unanalyzed data of the data unit data remains. If it remains, the process proceeds to S03, and if not, the process proceeds to S25.

  In S03, the first 1 byte is confirmed, and it is confirmed whether or not it is the start code of the control code. If it is not the start code of the control code, it is determined as a character code, and the process proceeds to S04. If it is the start code of the control code, the process proceeds to S05.

  In S04, if it is a 1-byte code, the character code of that byte is converted from an 8 unit code to a general character code such as UTF, and if it is a 2-byte code, the second byte is obtained and the 2-byte character code is obtained. Is converted from an 8-unit code into a general character code such as UTF, added to the acquired character string, and the process returns to S02.

  In S05, it is confirmed whether or not the start code is a start code of an extended control code (CSI). If the start code is not the extended control code (CSI), the process proceeds to S06. If the start code is the extended control code (CSI), the process proceeds to S07.

  In S06, in accordance with the syntax of each control code, a parameter having a predetermined number of bytes following the start code is acquired as a control code, and the process proceeds to S09.

  In S07, the data is acquired for each byte and is acquired until the byte becomes a terminal character, and is set as an extended control code, and the process proceeds to S08.

  In S08, it is determined whether the terminal character is auxiliary information for translation. If it is not auxiliary information for translation, the process proceeds to S09, and if it is auxiliary information for translation, the process proceeds to S10.

  In S09, the existing control code and extended control code are processed for the control code acquired in S06 and the extended control code acquired in S07, and the process returns to S02.

  In S10, it is determined whether or not the parameter of the auxiliary information for translation is the information for delimiter for translation execution. If it is not the delimiter information for translation execution, the process proceeds to S12, and if it is delimiter information for translation execution, the process proceeds to S11. Transition.

  In S11, the acquired character string stored so far is translated, the acquired character string is cleared, and the process returns to S02.

  In S12, it is determined whether or not the parameter of the auxiliary information for translation is information on the start of a name proper noun. If it is not information on the start of a name proper noun, the process proceeds to S14. Transition.

  In S13, the character added next to the acquired character string is set to be the starting character of the personal name proper noun, and the process returns to S02.

  In S14, it is determined whether or not the parameter of the auxiliary information for translation is information on the end of the name proper noun. If it is not information on the end of the name proper noun, the process proceeds to S16. Transition.

  In S15, in the acquired character string, the personal name proper noun start character set in S13 to the last character of the acquired character string are converted into a character string indicating a personal name proper noun described later, and the process returns to S02.

In S16, it is determined whether or not the parameter of the auxiliary information for translation is information on the start of a place name proper noun. If it is not information on the start of a place name proper noun, the process proceeds to S18. Transition.

  In S17, the character added next to the acquired character string is set as the start character of the place name proper noun, and the process returns to S02.

  In S18, it is determined whether or not the parameter of the auxiliary information for translation is information on the end of a place name proper noun. If it is not information on the end of a place name proper noun, the process proceeds to S20. Transition.

  In S19, in the acquired character string, the place name proper noun start character set in S17 to the last character of the acquired character string are converted into a character string indicating a place name proper noun described later, and the process returns to S02.

  In S20, it is determined whether or not the parameter of the auxiliary information for translation is other proper noun start information. If it is not other proper noun start information, the process proceeds to S22. Transition.

  In S21, the next character added to the acquired character string is set to be the start character of other proper nouns, and the process returns to S02.

  In S22, it is determined whether or not the parameter of the auxiliary information for translation is other proper noun end information. If it is not other proper noun end information, the process proceeds to S24, and if it is other proper noun end information, the process proceeds to S23. Transition.

  In S23, from the place name proper noun start character set in S21 in the acquired character string to the last character of the acquired character string are converted into character strings indicating other proper nouns described later, and the process returns to S02.

  In S24, the parameter of the auxiliary information for translation has a value outside the assumed range, so this extended control code is discarded and the process returns to S02.

  If it is determined in S02 that there is no remaining data unit data to be processed, the process proceeds to S25, and if data remains in the acquired character string, the data of the next subtitle PES is retained while retaining the data. The process proceeds to S26, and the process proceeds to S26 and the process ends.

  Next, a specific method for converting to a character string indicating a personal name proper noun in S15, converting to a character string indicating a place name proper noun in S19, or converting to a character string indicating another proper noun in S23 will be described.

  FIG. 14 is a chart showing data transition by the process using the information on the start of the name proper noun and the information on the end of the name proper noun. Specifically, it is a chart showing data transition when processing the code string shown in FIG. 11, which is a subtitle stream corresponding to (H) of one scene of a drama with a character named Hayashi in FIG. .

  The sequence (1) is a code sequence of the subtitle stream, followed by a control code sequence such as a display position and a display color, followed by a character code sequence of “it is”, and further followed by a control code for starting a name proper noun, Then, the character code string “Hayashi” continues, followed by the control code for the end of the name proper noun, followed by the character code string “I”, and finally the control code for the translation execution delimiter.

  (2) indicates that the result of extracting the character string from the code string of (1) is a character string “It is Hayashi”.

  (3) is an example in which the process from S15 in FIG. 13 to the character string indicating the personal name proper noun from the set personal name proper noun start character to the last character of the acquired character string is converted. Here, since the string “Hayashi” is likely to be translated as a common noun, it is converted to “Sato”, a proper noun for human names that is difficult to translate as a common noun. It is Sato. "

  (4) indicates that the character string as a result of translating the character string of (3) into English is “It is Sato.”.

  (5) indicates that the character name “san” is generated by adding the personal name proper noun “san” to the personal name proper noun “lin” before the conversion in (3). Yes.

  (6) indicates that the character string obtained by translating the character string of (5) into English is “Mr. Hayashi.”.

  (7) deletes “Mr.”, which is the translation result of “san” added in (5) from (6), and “Hayashi” as a translation result corresponding to the personal name noun “Hayashi”. . ”Is obtained and replaced with the character string“ Sato. ”In the conversion result of (4) to obtain“ It is Hayashi. ”, Thereby including the personal name proper noun“ Lin ”. "It is Hayashi" indicates that the character string to be output as a translation result is generated.

  In this case, “Sato” was used as the character string indicating the name proper noun, but this character string is inappropriate as the character string indicating the proper noun of the place name, and “Tokyo” as the character string indicating the place name proper noun. and so on. In the case of a name that is neither a personal name nor a place name, such as a building name, an organization name, or a company name, there is “Shinjuku Gyoen”.

  Here, “Sato”, “Tokyo”, and “Shinjuku Gyoen” are described as examples of character strings indicating respective proper nouns, but the present invention is not limited to these examples.

In addition, if the translation function evolves in the future and a specific symbol (for example, <and>) is added before and after the proper noun, it will be judged as a proper noun and it will have a function to translate it. The character string indicating the proper noun may be a character string provided with specific symbols before and after.
[Embodiment 2]
FIG. 15: is a block diagram which shows schematic structure of an example of the content reproduction apparatus with a caption concerning the 2nd Embodiment of this invention. A difference from the block diagram of the first embodiment in FIG. 1 is that a translation server communication unit 201 is added in place of the translation execution unit 109 so as to connect to a translation server 2002 outside the content playback apparatus 2001 with captions. It is a point.

  The translation server communication unit 201 uses the translation auxiliary information acquired by the translation auxiliary information acquisition unit 107 to translate the subtitle character string acquired by the subtitle character string acquisition unit 106 by the translation language determination unit 108. This is a part that requests an external translation server 201 to translate to the previous language and obtains the result of translation by the translation server 201.

The translation function requires a certain amount of processing power. By adopting such a configuration, an external server can be used as long as a function capable of connecting to the Internet is incorporated even in a device having a low processing capacity such as a TV. Translation can be realized.
[Embodiment 3]
FIG. 16: is a block diagram which shows schematic structure of an example of the content reproduction apparatus with a caption concerning the 3rd Embodiment of this invention. The content playback apparatus with subtitles 3001 is different from the block diagram of the first embodiment in FIG. 1 in that a translated character string reading unit 301 is added instead of the translated character string output unit 112.

  The translation character string reading unit 301 is a part that converts the translation result generated by the translation execution unit 109 into speech and outputs it. At this time, it may be mixed with the output sound output from the sound output unit 111 or may be output to a sound output device different from the sound output device such as a speaker output from the sound output unit 111.

In addition, the user may be able to select the speed and tone of voice reading in advance. Similarly, the user may be able to select a voice color in advance, but may be automatically determined as a voice included in the content and a voice that is easy to distinguish.
[Embodiment 4]
FIG. 17: is a block diagram which shows schematic structure of an example of the content reproduction apparatus with a caption concerning the 4th Embodiment of this invention. The content reproduction apparatus with subtitles 4001 is different from the block diagram of the first embodiment in FIG. 1 in that a translated character string transfer unit 401 is added instead of the translated character string output unit 112.

  The translation character string transfer unit 401 is a part that transfers the translation result generated by the translation execution unit 109 to a video output device different from the video output device such as a display output by the video output unit 110.

For example, the video output unit 110 may be displayed on a television (not shown), and the translated character string transfer unit 401 may be displayed on a nearby tablet terminal (not shown). In this case, make the character string of the translation result displayed on the tablet terminal at hand available to another application on the tablet, and use the copy and paste function to copy a part of the character string in the translation result. It may be possible to search on the Web.
[Summary]

  The first means of the present invention provides a subtitle analysis unit that extracts subtitle character strings and auxiliary translation information included in subtitled content, and translates based on the extracted subtitle character strings and translation auxiliary information. And a translation character string output unit that outputs a translation result generated by the translation execution unit as a character string.

  A second means is the first means, wherein the translation auxiliary information includes translation character string delimiter information indicating that a desired translation result can be obtained by translating with the subtitle character string extracted so far. It is characterized by including.

  A third means is characterized in that, in the first means, the auxiliary information for translation includes information on the start of a proper noun and the end of a proper noun.

  The fourth means is the third means wherein the start of the proper noun and the end of the proper noun are information on the start of the personal proper noun and the end of the personal proper noun, or information on the start of the place proper noun and It includes at least one of information on the end of a place name proper noun, information on the start of other proper nouns, and information on the end of other proper nouns.

  The fifth means converts the character string sandwiched between the start information of the proper noun and the start end information of the proper noun into a specific character string indicating a proper noun in the third means. And it is characterized by translating.

  The present invention can be used for all apparatuses for reproducing content with subtitles.

DESCRIPTION OF SYMBOLS 1001 ... Content reproduction apparatus 101 with subtitles ... Content receiving part 102 ... Stream separation part 103 ... Video stream decoding part 104 ... Audio stream decoding part 105 ... Subtitle stream decoding part 106 ... Subtitle character string acquisition part 107 ... Auxiliary information acquisition part for translation 108 ... Translation language determination unit 109 ... Translation execution unit 110 ... Video output unit 111 ... Audio output unit 112 ... Translation character string output unit 11 ... PES header 12 ... PES data 13 ... Data group header 14 ... Data group data 15 ... CRC
16 ... Subtitle sentence data header 17 ... Data unit 18 ... Data unit header 19 ... Data unit data 20a ... Control code string 20b ... Control code string 21a ... Character code string 21b ... Character code string 2001 ... Subtitled content reproduction apparatus 201 ... Translation Server communication unit 3001 ... Subtitled content playback device 301 ... Translated character string reading unit 4001 ... Subtitled content playback device 401 ... Translated character string transfer unit

Claims (5)

In a content playback device with subtitles that acquires and plays back content with subtitles
A subtitle analysis unit that extracts subtitle character strings and auxiliary information for translation included in the content with subtitles;
A translation execution unit that performs translation based on the extracted subtitle character string and auxiliary information for translation;
A content reproduction apparatus with subtitles, comprising: a translation character string output unit that outputs a translation result generated by the translation execution unit as a character string. In the content reproduction apparatus with subtitles according to claim 1,
The content reproduction apparatus with subtitles, wherein the auxiliary information for translation includes translation character string delimiter information indicating that a desired translation result can be obtained by translating with the subtitle character string extracted so far. In the content reproduction apparatus with subtitles according to claim 1,
The content reproduction apparatus with captions, wherein the auxiliary information for translation includes information on the start of proper nouns and the end of proper nouns. In the content reproduction apparatus with subtitles according to claim 3,
Information on the start of the proper noun and the end of the proper noun is information on the start of the proper name noun and the end of the proper name noun, or the start information on the proper name noun and the end of the proper name noun, or other proper nouns. A content reproduction apparatus with captions, comprising at least one of the start information and the end information of other proper nouns. In the content reproduction apparatus with subtitles according to claim 3,
With subtitles characterized in that a character string sandwiched between information on the start of the proper noun and information on the end of the start of the proper noun is converted into a specific character string indicating a proper noun and translated Content playback device