Replay device and method with automatic sentence segmentation
Contents
1. 5210 Receive a soundwave signal 220 Analyze the soundwave signal to determine plural segmentation points 5230 Receive a replay command and set a replay endpoint on the soundwave signal 5240 Set the nearest segmentation point prior to the replay endpoint as a replay startpoint S250 Play the soundwave signal from the replay startpoint to the replay endpoint FIG 2 U S Patent May 3 2011 Sheet 3 of 4 US 7 936 884 B2 S230 S242 Trace backwards to find the last segmentation point and set as the replay startpoint S244 Yes Determine if any replay command is not processed ye U S Patent May 3 2011 Sheet 4 of 4 US 7 936 884 B2 gt 250 S202 Determine if the replay should be stopped playing US 7 936 884 B2 1 REPLAY DEVICE AND METHOD WITH AUTOMATIC SENTENCE SEGMENTATION CROSS REFERENCES TO RELATED APPLICATIONS This non provisional application claims priority under 35 U S C 119 a on Patent Application No s 95145972 filed in Taiwan R O C on 2006 Dec 8 the entire contents of which are hereby incorporated by reference BACKGROUND OF THE INVENTION 1 Field of Invention The present invention relates to a replay device and method and more particularly to a replay device and method with automatic sentence segmentation function 2 Related Art Language learning apparatus or music player device cur2. rently has REPLAY function It facilitates user s repeated listening when learning a language or interested in some section of music thereby not wasting time on backward operations A device with general voice replaying function is only capable of manually setting time points A and B as a replay period and replay repeatedly Or setting the time point A replay startpoint first the device with general voice replay ing function will automatically calculate a delayed time period as the time point B replay endpoint However for user s actual operating situation when cer tain sentence or music section is unclear or user intends to listen again usually user already missed the desired time point A Then the backward operation needs to be applied to search for the expected time point A and set as replay start point Furthermore as a result of user s various responding speeds the set time point A is different from the expected time point of user and causes usage inconveniences Presently a player device is already provided with an auto matically setting function of replay start address point A the user does not have to set it manually The method is to set the current time point as the replay endpoint B when the user actuates the replay function and then trace a certain time period such as 10 seconds backwards to set as the replay startpoint However although such method no longer needs the user s manual setting operati
3. Receive a replay command and set a replay endpoint on the soundwave signal When the user is listening to the soundwave signal the replay command may be given at any time When the replay command is received there are three approaches to determine the replay endpoint on the soundwave signal first the replay endpoint is set as the time point on the soundwave signal when the user gives the replay command second trace backwards on the soundwave signal to find the end point of the last segmented sentence and set as the replay endpoint when receiving the replay command third trace forwards on the soundwave signal to find the end point of the current segmented sentence at the time point of receiving the replay command and set as the replay endpoint Step S240 Set the nearest segmentation point prior to the replay endpoint as a replay startpoint According to the replay endpoint set by step S230 step S240 traces backwards to find the nearest segmentation point prior to the replay endpoint as the replay startpoint Step 250 Play the soundwave signal from the replay startpoint set by step 240 to the replay endpoint set by step 230 Please refer to FIG 3 which shows a detailed flow chart for setting a replay startpoint according to an embodi ment of the present invention Step S240 is the step for setting the replay startpoint which may be subdivided to the follow ing steps Step S242 Trace backwards to find the last segmentation point
4. The method of claim 8 wherein the playing step plays the soundwave signal from the replay startpoint to the replay endpoint for multiple times
5. of every sentence The replay actuation module 20 receives the replay com mand and determines the replay endpoint on the soundwave signal The replay command is given by the user When the user is listening to the soundwave signal the replay command may be given at any time to notify the replay device in the present invention of the replay operation When the replay actuation module 20 receives the replay command there are three approaches to determine the replay endpoint on the soundwave signal The first approach is that the replay end point is set as the time point on the soundwave signal when the user gives the replay command Therefore when the replay actuation module 20 receives the replay command the replay endpoint will be set on the soundwave signal received by the soundwave analysis module 10 and set as the time point of receiving the replay command The second approach is to trace backwards on the sound wave signal to find the end point of the last segmented sen tence and set as the replay endpoint when the replay actuation module 20 receives the replay command Accordingly the user is allowed listening to complete sentence s when actu ating the replay function to play again The third method is to trace forwards on the soundwave signal to find the end point of the current segmented sentence at the time point of receiving the replay command and set as the replay endpoint when the replay actuation module 20 receives the repl
6. tion DETAILED DESCRIPTION OF THE INVENTION Please refer to FIG 1 which shows an explanatory dia gram for a replay device with automatic sentence segmenta tion The replay device with automatic sentence segmentation US 7 936 884 B2 3 includes a soundwave analysis module 10 a replay actuation module 20 a replay access module 30 and a replay module 40 The soundwave analysis module 10 receives the sound wave signal the soundwave signal may be voice declamation such as English conversation for a user to learn listening In the other hand when the user is listening to the voice decla mation sometimes there exits background sound within For example when listening to Japanese conversation there might be gentle orchestral music accompanying within as the background sound Therefore the soundwave signal may be the voice declamation plus the background sound After the soundwave analysis module 10 receives the soundwave signal it analyzes the soundwave signal to deter mine plural segmentation points Since the voice declamation is played sentence by sentence and between each sentence is so called sentence segmentation declaiming an article or playing a section of voice will generate plural segmentation points To determine the segmentation points for every sen tence may allow the user to trace backwards to the start point of every voice sentence when listening the playing of the soundwave signal and intending to replay and l
7. US007936884B2 a2 United States Patent 10 Patent No US 7 936 884 B2 Yen et al 45 Date of Patent May 3 2011 54 REPLAY DEVICE AND METHOD WITH 56 References Cited AUTOMATIC SENTENCE SEGMENTATION U S PATENT DOCUMENTS 75 Inventors Jui Yu Yen Kaohsiung TW Ming 5 231 239 A 7 1993 Tsumura et al 84 609 Hsiang Yen Kaohsiung TW Yi Chin 5 827 988 A 10 1998 Wachi cscs 84 609 Lin Fongshan TW 6 316 710 B1 11 2001 Lindemann ccss0 84 609 6 678 661 B1 1 2004 Smith etal 00 lt 704 278 3k I3 Assignee Micro Star International Co Ltd 7187290 B2 3 2007 Hydeetal o 340 5721 Jung He TW 7 668 434 B2 2 2010 Mitsuyu sissoo 386 219 2004 0114475 A1 6 2004 Suzuki etal 369 30 08 Notice Subject to any disclaimer the term of this 2008 0140391 A1 6 2008 Yen et al sees 704 200 patent is extended or adjusted under 35 2010 0011293 Al 1 2010 Yangetal 715 716 U S C 154 b by 1105 days FOREIGN PATENT DOCUMENTS TW 1255412 5 2006 21 Appl No 11 676 219 cited by chamines 22 Filed Feb 16 2007 Primary Examiner Xu Mei 74 Attorney Agent or Firm Apex Juris pllc Tracy M 65 Prior Publication Data Hiems US 2008 0140237 Al Jun 12 2008 57 ABSTRACT 30 Foreign Application Priority Data A replay device and method thereof with automatic sentence segmentation is provided A soundwave analysis module Dec 8 2006 TW viececceecssecsteseesneenes 95145972 U receives a soundwave si
8. and set as the replay startpoint Step 244 Determine if any replay command is not pro cessed yet If there is any replay command not processed yet go back to step S242 If there is no replay command not processed go to step 250 In step S230 if the replay command is continuously received for multiple times first trace backwards from the replay endpoint to find the last segmentation point through step S242 and set as the replay startpoint Next step S244 will determines if there is any replay command not processed yet Since the replay command is received for multiple times go to step 242 and trace backwards from the previously set replay startpoint to another former segmentation point Then the one before the last segmentation point will be reset as the replay startpoint until the replay command s is processed Therefore if the user wants to return to the N segmentation point prior to the replay endpoint simply gives the replay command for N times Then the replay startpoint will be set as any desired segmentation point Accordingly even though in the beginning the last segmen tation point that step S242 automatically traces back is not the desired replay startpoint the user may return to the one before 0 jak 5 40 45 50 55 65 6 the last segmentation point by repeatedly giving the replay command one more time By means of giving the replay command for multiple times the user may actually retur
9. arest segmentation point prior to the replay endpoint as the replay startpoint The replay module plays the soundwave signal from the replay startpoint set by the replay access module to the replay end point set by the replay actuation module Further scope of applicability of the present invention will become apparent from the detailed description given herein after However it should be understood that the detailed description and specific examples while indicating preferred embodiments of the invention are given by way of illustration only since various changes and modifications within the spirit and scope of the invention will become apparent to those skilled in the art from this detailed description BRIEF DESCRIPTION OF THE DRAWINGS The present invention will become more fully understood from the detailed description given hereinbelow illustration only and thus are not limitative of the present invention and wherein FIG 1 is an explanatory diagram for a replay device with automatic sentence segmentation according to an embodi ment of the present invention FIG 2 is a flow chart for a replay method with automatic sentence segmentation according to an embodiment of the present invention FIG 3 is a detailed flow chart for setting a replay startpoint according to an embodiment of the present invention FIG 4 is a detailed flow chart for playing a replayed sound wave signal according to an embodiment of the present inven
10. ation points what the user hears while replaying will always be starting from the beginning of each sentence which helps the user to listen complete sentences A replay device with automatic sentence segmentation pro vided in an embodiment of the present invention includes a soundwave analysis module a replay actuation module a replay access module and a replay module The soundwave analysis module receives the soundwave signal and analyzes the soundwave signal to determine plural segmentation points To determine the segmentation point there may be three methods First analyze the soundwave energy of the soundwave signal second use spectrum analy sis and third apply voice identification The replay actuation module receives a replay command and set a replay endpoint on the soundwave signal When receiving the replay command there are three ways to set the replay endpoint on the soundwave signal first set the time point corresponding to the soundwave signal of receiving the replay command as the replay endpoint second at the time point of receiving the replay command trace backwards on the soundwave signal to find the end point of the last seg mented sentence and set as the replay endpoint third when receiving the replay command trace forwards on the sound wave signal to find the end point of the current segmented sentence with the replay command received and set as the replay endpoint The replay access module sets the ne
11. ay command Such approach enables the user to listen to complete sentence s as well when actuating the replay function to play again The differences from the former approach described above is that the second approach traces backwards on the soundwave signal to find the last seg mented sentence but the third approach traces forwards on the soundwave signal to find the end point of the current segmented sentence wherein the above three setting approaches for the replay endpoint may be selectable by demand for the user The replay access module 30 sets the nearest segmentation point prior to the replay endpoint as the replay startpoint According to the replay endpoint set by the replay actuation module 20 the replay access module 30 traces backwards to find the nearest segmentation point prior to the replay end point and set as the replay startpoint If the user continuously gives the replay commands for multiple times that means the user intends to trace backwards to the former and former segmentation point So when the replay actuation module 20 receives the replay commands for multiple times the replay access module 30 will reset the one before the last segmen tation point as the replay startpoint Consequently even the former one before the last segmentation point that the replay access module 30 automatically traces backwards is not what the user wants the replay command may still be given repeat edly for mult
12. gnal and analyzes the soundwave signal to determine plural segmentation points A replay 51 Int Cl actuation module receives a replay command and sets a ais replay endpoint on the soundwave signal A replay access HOIR Solana 2006 01 module sets the nearest segmentation point prior to the replay 52 Us Cle ise soe Cate 381 56 700 94 463 35 endpoint as a replay startpoint A replay module plays the 58 Field of Classification Search 381 56 soundwave signal from the replay startpoint to the replay 381 58 119 110 700 94 463 35 84 604 84 609 704 207 208 210 214 217 270 272 See application file for complete search history START endpoint 14 Claims 4 Drawing Sheets 210 Receive a soundwave signal 220 Analyze the soundwave signal to determine plural segmentation points Receive a replay command and set a replay endpoint on the soundwave signal Set the nearest segmentation point prior to the replay endpoint as a replay startpoint Play the soundwave signal from the replay startpoint to the replay endpoint 230 240 250 U S Patent May 3 2011 Sheet 1 of 4 US 7 936 884 B2 soundwave signal 10 soundwave analysis module replay command 30 replay actuation replay access module module 40 replay module FIG 1 U S Patent May 3 2011 Sheet 2 of 4 US 7 936 884 B2 START
13. iple times to reach the desired replay startpoint Eventually the replay module 40 plays the soundwave signal from the replay startpoint set by the replay access module 30 to the replay endpoint set by the replay actuation module 20 The time that the replay module 40 plays the soundwave signal from the replay startpoint to the replay endpoint may be one time or multiple times which may be selectable by demand for the user US 7 936 884 B2 5 Please refer to FIG 2 which shows a flow chart fora replay method with automatic sentence segmentation according to an embodiment of the present invention The replay method includes the following steps Step 210 Receive a soundwave signal which may be the voice declamation only or the voice declamation plus the background music Step 220 Analyze the soundwave signal to determine plural segmentation points Analyzing to determine the seg mentation point may be achieved through the following three methods The first method is to analyze the soundwave energy of the soundwave signal The segmentation point will be determined when the soundwave energy of the soundwave signal is lower than an average value The second method is to use spectrum analysis to differentiate the soundwave signal between voice declamation and background music and deter mine the segmentation point of the voice declamation The third method is to use voice identification to determine the segmentation point Step 230
14. isten again Therefore to ensure the user to actuate a replay function may trace backwards to the start point of the last sentence or any of the former sentences and restart to listen The soundwave analysis module 10 keeps determining the segmentation point s when the soundwave signal is played normally Accordingly the user may listen to the normal playing of the soundwave signal and the soundwave analysis module 10 of the present invention analyzes the segmentation point s at the same time For the soundwave analysis module 10 analyzing the soundwave signal and determine plural segmentation points may be achieved through the following three methods The first method is to analyze the soundwave energy of the sound wave signal The segmentation point will be determined when the soundwave energy of the soundwave signal is lower than an average value When playing a section of the soundwave signal the soundwave energy has highs and lows From the user s aspect that sounds like so called special cadence in voice Briefly the voices thereof will have differences between loud and soft or high and low For voice declama tion after ends a sentence and before starts the next there is a short mute period or the soundwave energy decreases which is so called the segmentation point The first method is to use such characteristic First we set an average value for the soundwave energy when playing the soundwave signal the segmentation point
15. l from the replay startpoint to the replay endpoint is multiple times 8 A replay method with automatic sentence segmentation comprising the steps of receiving a soundwave signal analyzing the soundwave signal to determine a plurality of segmentation points US 7 936 884 B2 7 receiving a replay command and setting a replay endpoint on the soundwave signal setting a nearest segmentation point prior to the replay endpoint as a replay startpoint and playing the soundwave signal from the replay startpoint to the replay endpoint 9 The method of claim 8 wherein the analyzing step analyzes the soundwave energy of the soundwave signal the segmentation point being determined when the soundwave energy of the soundwave signal is lower than an average value 10 The method of claim 8 wherein the analyzing step uses spectrum analysis to differentiate the soundwave signal 8 between a voice declamation and a background music and determine the segmentation point of the voice declamation 11 The method of claim 8 wherein the analyzing step uses voice identification to determine the segmentation point 12 The method of claim 8 wherein the one before the last segmentation point is reset as the replay startpoint when the replay command is continuously received for multiple times 13 The method of claim 8 wherein the playing step plays the soundwave signal from the replay startpoint to the replay endpoint for one time 14
16. may be determined as long as the sound wave energy is lower than the average value The second method is that the soundwave analysis module 10 uses spectrum analysis to differentiate the soundwave signal between voice declamation and background music and determine the segmentation point of the voice declama tion As mentioned above sometimes the soundwave signal has the voice declamation and the background music exiting at the same time Since there are different spectrum between the voice declamation and the background music the sound wave analysis module 10 may uses spectrum analysis to dif ferentiate between the voice declamation and the background music Then use the first method or other methods to deter mine the segmentation point on the differentiated voice dec lamation The third method is that the soundwave analysis module 10 uses voice identification to determine the segmentation point The objective of voice identification is to enable electrical apparatuses to acknowledge human voices to perform corre sponding tasks When the testing voice is transferred from 0 an 5 25 30 40 45 65 4 analog to digital and then input and stored digitally the voice identification program will use a voice sample stored in advance to compare with the input testing voice sample Therefore through voice identification every sentence in the soundwave signal can be identified to determine the segmen tation point s
17. n points a replay actuation module receiving a replay command and setting a replay endpoint on the soundwave signal a replay access module setting a nearest segmentation point prior to the replay endpoint as a replay startpoint and a replay module playing the soundwave signal from the replay startpoint to the replay endpoint 2 The replay device of claim 1 wherein the soundwave analysis module analyzes the soundwave energy of the soundwave signal the segmentation point being determined when the soundwave energy of the soundwave signal is lower than an average value 3 The replay device of claim 1 wherein the soundwave analysis module uses spectrum analysis to differentiate the soundwave signal between a voice declamation and a back ground music and determine the segmentation point of the voice declamation 4 The replay device of claim 1 wherein the soundwave analysis module uses voice identification to determine the segmentation point 5 The replay device of claim 1 wherein the replay access module resets the one before the last segmentation point as the replay startpoint when the replay actuation module con tinuously receives the replay command for multiple times 6 The replay device of claim 1 wherein the time that the replay module plays the soundwave signal from the replay startpoint to the replay endpoint is one time 7 The replay device of claim 1 wherein the time that the replay module plays the soundwave signa
18. n to the desired one of replay startpoint Please refer to FIG 4 which shows a detailed flow chart for playing a replayed soundwave signal according to an embodi ment of the present invention Step S250 is the step for play ing the replayed soundwave signal which may include the following steps Step 252 Determine if the replay should be stopped play ing If No go back to step S250 and keep playing the soundwave signal from the replay startpoint to the replay endpoint If Yes go to End When the user feels like the replayed soundwave signal has been played for enough times just stop the replay may stop playing the soundwave signal from the replay startpoint to the replay endpoint Therefore the time for playing the sound wave signal from the replay startpoint to the replay endpoint may be one time or multiple times which is selectable by demand for the user The invention being thus described it will be obvious that the same may be varied in many ways Such variations are not to be regarded as a departure from the spirit and scope of the invention and all such modifications as would be obvious to one skilled in the art are intended to be included within the scope of the following claims What is claimed is 1 A replay device with automatic sentence segmentation comprising a soundwave analysis module receiving a soundwave sig nal and analyzing the soundwave signal to determine a plurality of segmentatio
19. on for the replay startpoint the replay time period is limited to a certain time period backwards Besides when the voice is playing it is possible that the first sentence lasts for only 5 seconds and the second sentence lasts for 10 seconds The playing time period has to be decided according to the length of the sentence which is not a fixed and certain time Therefore by means of tracing the certain time period backwards to set as the replay start point the device actually cannot reach the replay startpoint as the user s expectation nor reach the beginning of the sen tence for the user to listen a complete sentence repeatedly Consequently how to make the player device with voice replay function automatically set a correct replay startpoint without the user s manual operation and allow the user lis tening a complete sentence repeatedly has become a problem in urgent need of solution SUMMARY OF THE INVENTION Accordingly the present invention provides a replay device and method with automatic sentence segmentation which capable of automatically analyzing the segmentation points of a soundwave signal When the replay function is 20 25 30 35 40 45 50 55 60 65 2 actuated trace backward to find the nearest segmentation point and set as the replay startpoint Since the soundwave signal has been through the analysis of sentence segmentation in advance and all the set replay startpoints are segment
Download Pdf Manuals
Related Search