TWI233090B - System and method of language translation for multimedia data - Google Patents

Abstract

The present invention provides a system and method of language translation for multimedia data, which is to memorize the playing time and file name of the multimedia data during playing back of the multimedia data. Then record the memory data for the playing time according to the issued time recording instruction, and calculate to obtain a playing time range. Accordingly, fragment of audio data based on the playing time range for conversion is retrieved so as to obtain the necessary text data for display.

Description

1233090 V. Description of the invention (1) [Technical field to which the invention belongs] The present invention relates to a system and method for assisting Chinese language learning, particularly a system and method for language translation of multimedia materials. [Previous technology] Domestic investors have been able to expressly accept the interest of filmmakers. This service has been used as a direct source of information and the needs of the audio-visuals. If you want to understand its photos, it is more difficult to highlight the current situation. If you encounter the following At this time, due to the inquiries of the viewers, but the current situation of the function of the home video has been used for text messages, the video-style broadcast data is over, and the auxiliary language display that satisfies the concept of the word will be provided. It is necessary to watch several kinds of love movies to hold the current shadowing energy for query, but next, in the course of the process, it will take 5 reasons, and the inevitable media page of the father will be released, and the multimedia resources will be broadcasted, only for the viewers The study of film usage or classical language is to include the video contained in the multimedia materials to increase the provision of text messages in the local language or handicap materials, which does not meet the viewer's understanding of the content of the film. When using the relevant information of the French message, you can use the external tools to assist the learning. You can enjoy the video status: enjoy the continuous playback and have doubts about the dialogue. At the same time, you can play the movie because you have to watch and prepare more. It will be difficult to understand. The question is not provided when the video is provided. It can be missed, and all the viewers are troubled by a single language. Those who appreciate language learning in the video may encounter unrelated words and may have to watch the follow-up query to learn the language is not the other way. Can understand the inquiries of the self-provided words in the film can speak and want to work, all of the content provided by the dialogue to one The ministry 'will be able to talk to the internal tools and dictionaries; although the broadcast of the video will be paused while watching the video, the text message will be 1233090 V. Description of the invention (2) ^ Difficult to use word dictionaries and other auxiliary learning tools to understand the use of single words [ Summary of the Invention: In view of the problems of the above-mentioned conventional technology, the present invention provides a system and method for translation of body data into Chinese. Volume data = release time and multi-material release. Then record the heart according to the issued day-to-day record, record and play. Record = multimedia data file name, and then calculate a playback time.

The system and method for extracting the segment audio data into J :: J based on this playback time range to obtain the necessary text data for display. The purpose of the present invention is to provide a system and method for multi-dial translation. The purpose of this system and method is to perform language-assisted language learning. Therefore, the system of the present invention mainly includes a timing unit, a capture module, and an analysis conversion module. The system further includes a language interpretation module. "Koku" Among them, the operation method of the present invention mainly includes two types of real-time mode (Real Time Mode), ^ operation mode, one

Hoid Mode), the other is the capture mode ("Mode (On Mode (—1 丁 ⑽Mode) operation steps eVe Mode)" real-time instructions to record the playback time data 呓 悻. Jealousy% f. 100 first reception time record calculation Obtain the playback time range ^: Zhuo ;, the playback time memory data segment of the data is displayed once f, and the multimedia word data is displayed according to the playback time range display. Shell material; and finally converted to the text U fruit type (On) based on the audio data of the segment Hold Mode) and Retrieve Page 5 1233090 V. Description of Invention (3)

Mode) includes the following operating steps in common: · First receive the time recording instruction to record the playback time memory data and the multimedia data file name; then receive the query instruction and retrieve the multimedia data according to the multimedia data file name; then calculate and obtain the playback based on the playback time memory data Time range; and extract the audio data of the multimedia data according to the playback time range; and finally convert it into text data according to the audio data of the fragment. Through the foregoing brief description of the system and method disclosed by the present invention, the problems existing in the background description can be solved, and the following effects are expected to be achieved: 1 · Provide content query and translation functions, which can be used by users to solve problems Understand the question of dialogue; 2. Provide three ways to help users understand the content of dialogue, which can meet the various requirements of users when enjoying the video; ^ 3 · To ensure the accuracy of the displayed text data, a certain range can be achieved by tolerance calculation Processing of the audio data to improve the accuracy of the dialogue ^ ^; and 4. Allow the user to adjust the range of audio data capture forward or backward to meet the user's requirements for correct understanding of the dialogue content.

The features and implementation of the present invention are described in detail with the illustrations below: Fetching K 歹 M 歹 J [Embodiment] A system and method for language translation of multimedia data file names and body data for multimedia media playback , At the same time, memorize the playback time of the multimedia data at the same time and then record the broadcast according to the time recording instruction issued.

1233090 V. Description of the invention ⑷ Play time memory data and multimedia data files 幺 ^ Play time r Secret iP it μμ Connect I name ’and then calculate a broadcast i 隹 Μ ^ ^ Only then do you take out the audio information from the clip

Ding Qianjin converted to obtain the necessary text. Yin Nai only the S method, and the system and method of the display of Besanko 44 ΛΛ Τ in Beco are based on the operation of this system and method, which can be described as π ΛΑ. To achieve the above purpose, J has reached the goal of assisting Chinese language learning. Firstly, the explanation of the 97-key transmission between each unit module of the language auxiliary system of the present invention is carried out by "the first text auxiliary = diagram". It is shown that the language auxiliary system of the present invention mainly includes the replacement of the small / self group 110, the timing unit 120, the acquisition module 130, the analysis conversion = group UG, and the database 15. In addition, it can also include _ language interpretation modules

Jianshi among which the memory module 1 1 0 is responsible for memorizing playback time memory data and playback multimedia data '· 10 o'clock unit i 20, responsible for recording the elapsed time of multimedia data broadcast and providing playback time memory data; capture Module 13〇 is responsible for calculating the playback time range based on the playback time memory data, and = extracting the audio data of the multimedia data based on the playback time range; analysis ▲, module 1 4 0, responsible for comparing the data according to the audio fuzzy 1 5 1 Analyze the fragment audio data and convert the fragment audio data to obtain text data; the database says 50 0 'is responsible for storing a plurality of audio fuzzy comparison data 丨 51 and plural texts to explain the poor information; the language interpretation module 1 6 0 Is responsible for providing text interpretation information based on text data and text 'selection instructions. Among them, the staff removal module 132 (as shown in "Figure 1-b group 130 further includes a setting unit 131 and a calculation unit), the setting unit 1 3 1 is responsible for providing each 1233090

2 Set the operation interface to provide the change of the time acquisition parameters. [The cutting mode of the day surface is changed. As for the calculation unit 132, the mode, Ϊ = ί parameters, perform calculation operations to obtain the playback time range, and k for extraction. The branch group 丨 30 performs the subsequent operation of capturing the audio data of the segment 丨 in addition, the analysis conversion module 丨 40 also includes a sound loudness analysis module Η ^, an audio frequency analysis module 142, a fuzzy comparison module 143, and a voice file 1 (St. "Figure 1-C"), the acoustic loudness analysis module 141 and the = frequency analysis module 142 are responsible for jointly analyzing the loudness and frequency of the audio data, and through sound filtering to obtain a Pure audio, the module 2 comparison module 143 performs fuzzy comparison on the pure audio audio according to the audio fuzzy comparison data provided by the database 150 to obtain a human voice, and finally the voice is converted to a text mode. Group 44 is converted into text information based on this human voice to provide system display. When the user uses the system of the present invention to play a movie, the system first stores the multimedia data of the movie and stores the multimedia shell with a memory module 1 1 0; and the timing unit 1 2 at the same time as the playback. 0 Record the elapsed time of the multimedia data broadcast. When the system receives the capture command, it will use the capture mode and compare the time elapsed with the video playback according to the capture command. Time memory data, and record the playback time memory data in ^ Ji 丨 Think module 1 1 0, and then retrieve module 1 3 0 according to the translation playback mode selected by the user when f is opened, and according to the playback time Calculate the memory data to obtain the playback time range, and then compare the multimedia data based on the playback time range to extract the audio data of the clip for audio data analysis.

1233090 V. Description of Invention (6) '----- Industry. Secondly, after acquiring the audio data of the segment from the acquisition module 130, the audio data of the segment is transmitted to the analysis and conversion module 丨 4 for analysis and conversion operation, and the analysis and conversion module 14 from the database 1 50 Extract the audio fuzzy comparison data 2 1 and use this audio fuzzy comparison data 1 5 1 to perform fuzzy comparison and analysis on the captured segment audio data, convert the segment audio data into text data, and then Text information is displayed. After converting and obtaining the text data and displaying it, the user can also query the interpretation information of the text data for the displayed text information. At this time, the user's request f command, the system's language interpretation module 丨 6 According to this request, eighth, extract the textual interpretation information of this text data from Bayer Library 150 for users to further understand the content of the video dialogue. After describing the transmission status of the message discs between the units and modules in the system of the present invention, continue to use the second figure, the flow chart, and the third figure. Refers to ::; two of the three modes of operation. χ Translation of the invention of the invention ^ Provided 3 When the user chooses to use the real-time mode (Real view video, as shown in "Figure 2", 0de) to read media data, and at the same time broadcast, record multimedia data? \ Over time, while the video is playing, the player will correct the play order. At this time, the system continues to play the video and ^ 栌 protects the inter-recording record means = memory data (step 210); the system's capture mode Group 13〇: = After releasing the data of Nen Yi, then according to the obtained playback time 1233090 V. Description of the invention (7)

The data is calculated based on the IM range and the range (step 220); then the audio data of the snippet segment is converted by analysis during playback (step 230); and finally, the data center is the data analysis and conversion operation to convert to text. : The user chooses to use the pause mode (〇η "Μ M〇de) to read the" Capture mode (when watching the video, if it is more broadcast on Μ _ 糸), in addition to the multimedia data of the priority storage video,% Video playback 1 :: The elapsed time of the door body data playback, so that users do not need to find out the content of the dialogue in the video has instructions, the master can give the system time recording system, and the fragmented audio data. Time material and multimedia data file name (V step 32); Li Er; the mode of mode acquisition (data and multimedia data file name of Jt between Retrleve, that is, continue to play Γ 婢

Mode) Hold the name, that is, H Tongxi will get the playback time memory data and multimedia data slot: Retrieve the multimedia data according to the multimedia data file name (steps, ', and calculate the 〃 ^ of the film% based on the playback time memory data; , And then use this broadcast & time to capture the toast data 2: 2 ::::: material (step 340); finally, use the multimedia data piece ... to analyze and convert the operation, and to obtain the segment audio data turn

1233090 V. Description of the invention (8) Replaced with text data (step 3 5 0), and displays the obtained results. ^ = muetI * ieve MQde) m ^ After the release, the user first displays the playback material and multimedia data recorded by the user, and asks the user to select the asset data to retrieve and convert the segment audio data. After 2 sets of time-memory data are recorded, the system will then multiply according to the number :: select the name to retrieve the multimedia data (step 32), ·-4 to calculate the playback time range (step = = to retrieve the multimedia data) The segment audio data (step = broadcast; analysis and analysis of the segment audio data of the multimedia data. Turn 4: Take the segment audio data of t = and convert it into text data. The results obtained will be displayed at first glance.) They all include the steps of calculating and playing 0 Shimenzhangwei, and this step is composed of a series of "M「 series of operating processes, as shown in the figure, and with "V9 Figure 1 changing The process of the time range is said. I do not intend to solve this operation process according to Figure 9 "First, as shown in Figure 9", suppose that once you ’ve swallowed the β position (20:30:35), ^ Movies played to point B The white content produces an incomprehensible "use J;" For this point, the pair of movies received the time St :: cause in the system; a time recording instruction was issued, but the time position to run to point A: the time point of the movie playback may be The time recording instruction is at A point% 1 minute and 00 seconds) 'At this time, the system is based on 410); Assume that the time is retrieved and the time is stored to retrieve the data (steps ... The set time for taking the number of tea is 30 seconds At this time, the system is 1233090 V. Description of the invention (9) Reduction of time fetching loss according to the playback time memory data (step 420) 'that is, "Figure 9 ^ to get less playback time"; Set 2 rooms (20: 3. Minutes to get the end of the playback time (step 43)), that is, "add the location between the parameters (20:31:30); finally, I won't do it again." At the time point C in the middle, the playback time range is obtained (^ The starting point of Zhijian and the playback range are shown in Figure 4 as shown in Figure 4). The system length can be obtained by the operation flow. Take to assist Fan Fan after playing time Carry out the operation of the second time in the completion of the playback time range. You also need to capture the video according to the playback time range. 俨 = Wide, after that, the entire system industry can obtain the necessary text information ^ 9 information and convert it to Regarding the flow of the conversion operation, select = understand the dialogue content of the movie, the flow of the audio data of the fragment into the text data = Figure 5 "The conversion process of the present invention. After the audio data analysis of the sound loudness and sound frequency and oblique, that is, according to the loudness and frequency of the audio data of the clip ^ Step 510); in the way of obtaining the noise of the entire film, to obtain the film, continue to filter the audio 520)), then use the fuzzy comparison of $ ^ ° to clean the net audio audio (step fuzzy comparison of the data 151, and take ^ to the net audio audio with the audio blur ratio of 53 0); finally according to this Voice information service—personal voice information (step 540) to complete the conversion operation. Be converted to text data (steps through the above three browsing modes of% 1233090 V. After the description of the invention (10), after viewing the captured video and converted text times, if the user thinks about the displayed text information, Its correctness and accuracy ^, asks%, you can use the system to provide forward ^ ... ward) fetching means that the person is backward (Rewind) fetching instruction, reset the clip audio data capture / playback time range to get again A new piece of video and text data, the user confirms the correct dialogue content. In addition, there is a parameter about time acquisition = no, which is mainly set by the operating interface provided by the system. The operation surface is shown in "Figure 10". You can set the time acquisition parameter in minutes or seconds = unit, more Take the number as the size, and control the size of the playback time range to determine the amount of audio data of the clips. In addition, the system also provides a word query function, which can provide users with information about the text converted by the system ^ , To perform a word query to understand various explanations or reference examples of an unknown word, to help learners better understand the word. After describing the signal transmission methods and methods between the modules of the system of the present invention, After the basic operation process, continue to view the detailed operation flowchart of the embodiment by using the “Real Time Mode” of the present invention in “FIG. 6”, and view the embodiments by “0n Hold Mode” of the present invention in “FIG. 7” Detailed flowchart, "Figure 8" The detailed flowchart of the embodiment of the retrieval mode (Retrieve Mode) of the present invention, and the embodiment screens of "Figure 丨 丨" to "Figure 丨 3", Explanation of the preferred embodiment of the present invention. Before the user uses the system of the present invention to enjoy a video, the user can first set the value of the time capture parameter through the system's operation interface. After setting the Yuancheng, the system will use the time capture parameter to fade the playback time range. Based on, when the user chooses the browsing mode,

m

1233090 V. Description of the invention (11) Real Time Mode browsing mode Please refer to "Figure 6" for video viewing at the same time. The system receives the Real Time Mode browsing instruction (step 610) After that, the video will start to play on a single screen, as shown in "Figure 11"; during the playback of the video, when the system receives the time recording instruction (step 6 2 0); the system will then cut and display the daylight into a video And text display screen (step 63), and continue to play the video, as shown in "Figure 12" and "Figure 13", at this time the user can choose the way the cutting screen is displayed according to his preference; Then, because the viewing mode used by the system is Real Time Mode, after the system acquires time memory data according to the time recording instruction, in addition to continuously playing the shirt, it also calculates the playback time range based on the time memory data set by the first uranium. (Step 640); Then, according to the playback time, the information is converted and converted into text information (step 650); most ;; clerk = face text information (step) At this time, the two displayed are the text information obtained from the conversion of the audio data of the segment calculated at that time point when the time recording instruction is issued; then when the user reads & finishes, he can send an exit instruction through the system operation interface When the window is closed by 2, the day and time window is closed. At this time, after receiving the leave command (step 67), the system resumes the single display of the day and time window (step 680) and continues to play the movie. When the user selects the browsing mode, select the browsing mode for video viewing (2) to switch to video appreciation. ^ Change = Picture "The system first plays the video in a single-screen playback mode, such as" = After two videos are played for a period of time, when the user encounters something that is not clear in the dialogue between the movie and the movie, he sends out a pause show instruction:

1233090 V. Description of the invention (12) Ϊ t ί After receiving this pause browsing instruction (step 710); first pause the playing of the media tribute (step 72 0); then cut the display screen into video and text two f = day surface ( Step 725), as shown in "Figure 12" or "Figure 13", 'the person records the playback time memory data and the media media file name at the time point when the pause browsing instruction is received, and then receives the playback time After memorizing the data (step 730), calculate the playback time range (step 35); the system then extracts the audio data and converts it into text information based on the playback time range and the obtained multimedia data file name (step 740); and Display the text information on the daytime display (step 75); Finally, when the user finishes reading, 'send a leave command through the system's operation interface to close the double-view = ΐ plane, and the system is now receiving the leave command ( After step 760), in addition to restoring a single daytime display (step 77), the multimedia material will then continue to be played (step 780), and the movie content will continue to be played. 、 When the user chooses the browsing mode of the retrieval mode (Retrieve Mode) for movie appreciation, please refer to "Figure 8". At this time, the system first plays the movie in a single day-time playback mode, such as As shown in "Figure 11", after the video is played for a certain period of time, when the user encounters the content of the video dialogue that is not understood, the user sends a fetch command, and after receiving the hold command (step 810); that is, According to this fetch instruction " order, record and obtain the play time memory data (step 82), and record this play time memory data together with the multimedia data file name in the memory module, and continue to play the video after recording ; Then when the video finishes playing, the "operating system of the user" displays the stored time data of the stored multimedia data file (step 830); and then accepts the user for multimedia

Page 15 1233090 V. Description of the invention (13) Selection of data file name and playback time memory data (step 84); At this time, line ^ 切割 A is cut to display the day and time into video and text display screens (after step 2)选取 Calculate the playback time of the selected playback time memory data ^ = 6〇); and according to the multimedia data file name and step 870:: T%: Retrieve the audio data of the clip and convert it into text information. = ^ Yu: Vice video display $ 4 side broadcast 纟 the playback time range _ / meaning; 1 read and display text information on the text display screen (steps ask the user if the system converts the text information 'the system will ask for selection "Yes", scorn, despise (step 890) the recorded data. When the time is recorded, the system will display the saved multimedia data file name and broadcast the user. 83〇), provide the user to continue Inquiry and reading; learning assignments. No ", the system stops operating and the auxiliary language is not used to limit 2 =, it is only the preferred embodiment of this creation, and the scope of the implementation of the work ; That is, according to the original A patent-pending fan, culture and modifications are all covered by the scope of this patent creation.

1233090 A brief description of the diagram. Figure A is the structure diagram of the language auxiliary system of the present invention; Figure 1-B is a detailed composition diagram of the acquisition module of the language auxiliary system of the present invention; Figure 1-c is the invention The structure diagram of the analysis conversion module of the language-assisted system; Figure 1 and Figure 2 are the main flowchart of the instant translation of audio data according to the present invention;

FIG. 3 is a flowchart of retrieving audio data according to the retrieval instruction according to the present invention; J process diagram; FIG. 4 is a flowchart of converting and obtaining the playback time range according to the present invention; FIG. 5 is a converted segment audio of the present invention The flow of data into text data Figure 6 is a detailed operation flow chart of each example of the Real Time Mode browsing of the present invention; see Figure f 7 is a view of each of the On Hold Mode browsing of the present invention Detailed flow chart; Besch example is the RetrieVe Mode view of the actual Jδ flat flow chart; 焉 ^ 9 Figures / is a schematic diagram of the playback time range of the invention; Figures; The interface diagram for setting the time acquisition parameters in the embodiment of the present invention is a single-display screen diagram of the embodiment of the present invention; the diagram is the display of the embodiment of the present invention without intention; and Figure 3 is a schematic diagram of a simple illustration window with the display screen cut into left and right 1233090 drawings according to an embodiment of the present invention. [Illustration of Symbols]

Page 18 110 Memory module 120 Timing unit 130 Capture module 131 Setting unit 132 Calculation unit 140 Analysis conversion module 141 Acoustic degree analysis module 142 Sound frequency analysis module 143 Fuzzy comparison module 144 Sound conversion text mode Group 150 Database 151 Audio fuzzy comparison data 160 Language interpretation module

Claims (1)

1233090 VI. Scope of patent application 1. A system for language translation of multimedia data. It is a system that obtains text information by converting audio data of one of the multimedia data. The system includes: a memory module for Memorize a play time memory data; a timing unit to record the elapsed time of the multimedia data playback; an acquisition module to calculate a play time range based on the play time memory data and a time acquisition parameter, And extracting a piece of audio data of the multimedia data according to the playback time range; a database for storing a plurality of audio fuzzy comparison data; and an analysis conversion module for analyzing the audio fuzzy comparison data Segment audio data, and convert the segment audio data to obtain a text data. 2. The system for language translation of multimedia data as described in item 1 of the scope of the patent application, wherein the system further includes a language interpretation module that receives a text selection instruction to provide a text interpretation information based on the text data. 3. The language translation system for multimedia data as described in item 2 of the scope of patent application, wherein the text interpretation information is provided by the database 4. The language translation for multimedia data as described in item 1 of the scope of patent application System, the capture module further includes a setting unit for providing the time capture parameter. Page 19 1233090 VI. Application for patent scope 5 · Time range for applying for special translation 6 · Calculation for applying for special translation 7 · Oblique Ganshi as described in item 1 of the patent application scope The data is processed in the language system, and the time capture parameter J ^ J is used to change the playback: Fan; the language system for multimedia data described in item 1 above, where the capture module further includes a calculation unit, Use this playback time range. Take into account a range. 8 · If you want to apply for a special translation, you need to analyze the module. A voice is turned 9 · If you want to apply for a special translation, you need to analyze the module. Audio. 10 · If applying for a special translation is based on the audio. 11 · The system of language translation for multimedia data as described in item 1 of the scope of application for translation of monographs: where the capture module provides a forward (Reward 1) capture instruction, To change the language system for multimedia data as described in item 1 of the playback time range, where the analysis conversion module further includes a sound degree, a sound frequency analysis module, a fuzzy comparison module, and a text change mode. group. The language system for multimedia data as described in item 8 of the profit range is used, wherein the acoustic loudness analysis module and the sound frequency are used to analyze the audio data of the segment to obtain a target as described in item 9 of the net sound range. Multilingual data for language system 'Where the net audio and audio system provides the fuzzy comparison module to compare the fuzzy comparison data to obtain a speech system for multimedia data as described in item 10 of the vocal profit scope, 苴Φ Bee vocal information system provides the sound conversion text 1233090 Capture audio data according to the playback time range; and take a segment of the multimedia data According to the piece #, the audio data is converted into a text data. . The method of "Range No. 12 Institute" multimedia data. The method of acquiring the playback time range includes the following: receiving the playback time memory data; reducing the time acquisition parameter to obtain the playback time memory data to obtain A starting point of playing time; adding a time acquisition parameter according to the playing time memory data to obtain an ending point of playing time; and The playback time range is obtained according to the start of the playback time and the end of the playback time. 14 · The method for translating a language using multimedia data as described in item 丨 2 of the scope of patent application, wherein the step of converting the audio data of the segment into the text data further includes the following 2 steps: According to the audio information section of the segment Analyze a loudness and an audio 1233090, patent coverage rate; filtering a noise of the audio data of the segment to obtain a net audio; fuzzy comparison of the net audio and an audio fuzzy comparison data to obtain a voice information; and converting the voice information into the text data. 1 5. The method for translating languages using multimedia data as described in item 丨 2 of the scope of patent application, wherein the method is a real-time mode (Rea 1 ΐ 1 me Mode) browsing, which is based on receiving the time recording instruction While taking the audio data of the clip, continue to play the multimedia data. 16 · The method for translating languages using multimedia data as described in item 5 of the patent application scope, wherein the browsing of the Real Time Mode provides a forward fetch instruction and a backward (Rewind) ) Fetching instructions to change the playback time range. 1 7 · A method for language translation using multimedia data, which is a method of obtaining a text information by extracting a part of audio data in the multimedia data for conversion. The method includes the following steps: receiving a time recording instruction and recording a Play time memory data and a multimedia data file name; receive a query command and retrieve a multimedia data according to the multimedia data file name; calculate and obtain a play time range based on the play time memory data; retrieve the multimedia according to the play time range A piece of information Page 22, 1233090 VI. Patent application scope Audio data according to 18 · If applying for translation of monographs: Receiving based on obtaining a broadcast, obtaining obtaining a broadcast according to the time frame 19 · If applying for translation of textual data based on monographs; filtering news; Blur: And the audio data of the clip is converted into the profit method described in item 17 of the profit range, wherein the playback time is text data. The acquisition of the range using multimedia data includes the following: Playback Playback Playback Time Playback Playback Time Playback Range. The method of profit range, the steps of the time memory data of the segment; the time memory data minus the starting point; the time memory data increase the end point; and the time starting point and the benefits described in item 17 of this broadcast, which according to the film, also includes the following less one Time capture parameter to add a time capture parameter to release the time end to obtain the broadcast multimedia data to convert the segment audio data into the conversion step: audio data analysis one sound and one audio noise of the segment audio data To obtain a net tone-to-tone ratio of the net tone audio and one tone to obtain a vocal information; and the vocal information into the profit method described in item 17 of the textual profit scope, wherein the method is to convert 2 〇 · 如Obtain fuzzy comparison information for translation of monographs. Using multimedia materials to speak-On Hold mode Page 23 1233090 VI. The scope of patent application mode) is to stop playing the multimedia data while receiving the time recording instruction to retrieve the segment audio data. 2 1 · The method of multimedia translation for language translation as described in item 20 of the scope of patent application, wherein the browsing in the On Hold Mode provides a forward fetch instruction and a backward (Rewind) ) Fetch command to change the playback time range. 2 2 · The method of multimedia translation for language translation as described in item 7 of the scope of patent application, wherein the method is a retrieval mode (Retrieve Mode) browsing, which is recorded while receiving the time recording instruction The multimedia data file name and the playback time memorize the data, and continue to play the multimedia data. 2 3 · The method of multimedia translation for language translation as described in item 22 of the scope of the patent application, wherein the browsing of the Retrieve Mode provides a forward fetch instruction and a rewind Fetch command to change the playback time range. 2 4 · The method for multimedia translation using multimedia data as described in item 17 of the scope of patent application, wherein the method is a retrieval mode (Retrieve Mode) browsing, and the query instruction is provided based on the playback of the multimedia data. . ' Page 24