WO2016095280A1 - 卡拉ok评分方法和装置 - Google Patents
卡拉ok评分方法和装置 Download PDFInfo
- Publication number
- WO2016095280A1 WO2016095280A1 PCT/CN2014/095645 CN2014095645W WO2016095280A1 WO 2016095280 A1 WO2016095280 A1 WO 2016095280A1 CN 2014095645 W CN2014095645 W CN 2014095645W WO 2016095280 A1 WO2016095280 A1 WO 2016095280A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- pronunciation
- singer
- time
- frequency
- preset
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10H—ELECTROPHONIC MUSICAL INSTRUMENTS; INSTRUMENTS IN WHICH THE TONES ARE GENERATED BY ELECTROMECHANICAL MEANS OR ELECTRONIC GENERATORS, OR IN WHICH THE TONES ARE SYNTHESISED FROM A DATA STORE
- G10H1/00—Details of electrophonic musical instruments
- G10H1/36—Accompaniment arrangements
Definitions
- the present invention relates to the field of media broadcasting, and more particularly to an invented karaoke scoring method and apparatus.
- KTV karaoke scoring devices such as smart TV
- Sales are climbing, but when the singer sings KTV songs using a microphone, the delay in the sound caused by the delay in the transmission is delayed, and the singer's voice delay causes the karaoke scoring device to score the software, causing the score to be incorrect.
- the singer turned on the Scorpio K song application and selected the song "Love You for Ten Thousand Years" by Hong Kong and Taiwan star Andy Lau.
- the singer picked up the microphone and started singing against the lyrics due to the karaoke scoring device scoring system.
- the score displayed on the screen of the karaoke scoring device was 0, causing confusion for the singer. If the delay in the transmission of the singer's pronunciation is solved, the erroneous judgment of the karaoke scoring device scoring software is an urgent problem to be solved.
- the main object of the present invention is to solve the problem of erroneous judgment of the karaoke scoring device scoring software caused by the delay of the vocalist's pronunciation during transmission.
- the invention provides a karaoke scoring method, and the karaoke scoring method comprises the following steps:
- the vocalist's pronunciation time is calibrated according to the delay difference of the singer's pronunciation
- a score is given by comparing the pronunciation of the vocalist's calibrated pronunciation time with the pronunciation of the original singer.
- the step of comparing the pronunciation of the vocalist's calibration pronunciation time with the pronunciation of the original singer, and the step of giving a score further includes:
- the pronunciation frequency of the singer and the pronunciation frequency of the original singer are acquired while acquiring the pronunciation time; the pronunciation corresponding to the calibrated pronunciation time of the singer The frequency is compared with the pronunciation frequency of the original singer, and if there is a difference, the pronunciation frequency of the singer is compensated.
- the step of obtaining the vocalist's pronunciation time and the original singer's pronunciation time includes:
- the step of determining whether there is a vocalist's pronunciation, if any, acquiring the pronunciation time and the pronunciation frequency, and simultaneously obtaining the pronunciation time and the pronunciation frequency of the original singer specifically includes:
- the step of determining whether the system delay or the singer's own delay is based on the calculated delay difference of the singer's pronunciation includes:
- the present invention further provides a karaoke scoring apparatus, the karaoke scoring apparatus comprising:
- a obtaining module for obtaining a vocalist's pronunciation time and an original singer's pronunciation time
- a calculating module configured to calculate a delay difference of the singer's pronunciation according to the singer's pronunciation time and the original singer's pronunciation time
- a judging module for judging whether the system delays or the singer's own delay based on the calculated delay difference of the singer's pronunciation
- a calibration module for calibrating the vocalist's pronunciation time according to a delay difference of the singer's pronunciation if it is a system delay
- a scoring module for comparing the pronunciation of the vocalist's calibrated pronunciation time with the pronunciation of the original singer, giving a score.
- the karaoke scoring device further comprises:
- a frequency compensation module configured to acquire a pronunciation frequency of the singer and a pronunciation frequency of the original singer, the pronunciation frequency of the singer and the pronunciation frequency of the original singer are acquired while acquiring the pronunciation time; the singer's The pronunciation frequency corresponding to the calibration pronunciation time is compared with the pronunciation frequency of the original singer, and if there is a difference, the pronunciation frequency of the singer is compensated.
- the karaoke scoring device further comprises:
- the preset module is configured to determine whether there is a singer's pronunciation, and if so, obtain the vocalist's pronunciation time and pronunciation frequency, and obtain the original singer's pronunciation time and pronunciation frequency.
- the preset module comprises:
- An acquisition unit for collecting continuous pronunciation signals for collecting continuous pronunciation signals
- a conversion unit configured to convert the pronunciation signal into a digital signal
- An obtaining unit configured to acquire the number of times of pronunciation in the set time according to the converted digital signal
- a comparison unit configured to compare the number of pronunciations in the acquired set time with a preset pronunciation sub-interval, if the number of pronunciations in the acquired set time is within a preset number of pronunciation times, It is confirmed that the singer is singing, and if the number of times of the conversion is not within the preset number of times, it is confirmed that no singer is singing, and the pronunciation time and pronunciation frequency are no longer acquired.
- the determining module is configured to compare a delay difference of the singer's pronunciation with a preset time threshold, and if less than the preset threshold, determine that the system is delayed, according to the singer's pronunciation.
- the delay difference is used to calibrate the vocalist's pronunciation time; if it is greater than the preset threshold, it is determined that the singer himself is delayed, and the delay difference of the singer's pronunciation is not calibrated.
- the karaoke scoring method obtaineds the vocalist's pronunciation time and the original singer's pronunciation time; and calculates the singer's pronunciation delay difference according to the singer's pronunciation time and the original singer's pronunciation time. According to the calculated delay difference of the singer's pronunciation, whether the system delay or the singer's own delay; if it is a system delay, the vocalist's pronunciation time is calibrated according to the singer's pronunciation delay difference; The pronunciation of the vocalist's calibrated pronunciation time is compared with the pronunciation of the original singer, and a score is given.
- the invention improves the false positive rate of the karaoke scoring device scoring software, enhances the user experience, and increases the selling point of the karaoke scoring device.
- FIG. 1 is a schematic flow chart of a first embodiment of a karaoke scoring method according to the present invention
- FIG. 2 is a schematic flow chart of a second embodiment of a karaoke scoring method according to the present invention.
- FIG. 3 is a schematic flow chart of a third embodiment of a karaoke scoring method according to the present invention.
- FIG. 4 is a schematic diagram showing the refinement flow of the steps of determining the pronunciation time and the pronunciation frequency of the singer and obtaining the pronunciation time and the pronunciation frequency of the original singer, as described in FIG. 3;
- FIG. 5 is a schematic flowchart diagram of a fourth embodiment of a karaoke scoring method according to the present invention.
- FIG. 6 is a schematic diagram of functional modules of a first embodiment of a karaoke scoring apparatus according to the present invention.
- FIG. 7 is a schematic diagram of functional modules of a second embodiment of a karaoke scoring device of the present invention.
- FIG. 8 is a schematic diagram of functional modules of a third embodiment of a karaoke scoring apparatus according to the present invention.
- FIG. 9 is a schematic diagram of functional modules of the preset module in FIG. 8.
- the present invention provides a karaoke scoring method.
- the karaoke scoring method includes the following steps:
- Step S100 obtaining the vocalist's pronunciation time and the original singer's pronunciation time.
- the karaoke scoring device After detecting the vocal signal of the singer and the pronunciation signal of the original singer, the karaoke scoring device respectively obtains the vocalist's pronunciation time and the original singer's pronunciation time, wherein the singer's pronunciation time is played in the karaoke scoring device.
- the voice of the singer is not the time when the singer actually starts singing, because the singer will have a delay in the transmission of the microphone while singing, and the original singer’s pronunciation time is the singer’s song, the on-demand song
- the time when the Karaoke scoring device started playing the Andy Lau songs was the pronunciation time of the original singer.
- the karaoke scoring device can be a smart TV, a DVD player, a computer, a smart phone, a tablet, and the like.
- Step S200 Calculating a delay difference of the singer's pronunciation according to the singer's pronunciation time and the original singer's pronunciation time.
- the karaoke scoring device calculates a delay difference of the singer's pronunciation according to the obtained singer's pronunciation time and the original singer's pronunciation time, and the time difference is the original singer's pronunciation time minus the singer's pronunciation.
- the difference in time, the vocalist's pronunciation time is not based on the actual time, but from the start time of the Kali OK song, for X points X
- the format of the second for example, the vocalist's pronunciation time is 1 minute and 20 seconds, while the original singer's pronunciation time is 1 minute and 0 seconds, then the singer's pronunciation has a delay of 1 minute and 20 seconds minus 1 minute and 0 seconds.
- the result is 20 seconds, that is, the singer's pronunciation has a delay of 20 seconds.
- Step S300 determining whether the system delays or the singer's own delay according to the calculated delay difference of the singer's pronunciation.
- the karaoke scoring device initially judges whether the system delays or the singer's own delay according to the calculated delay difference of the singer's pronunciation, for example, compares the singer's pronunciation delay difference with a preset time threshold, if less than Determining a preset threshold, determining that it is a system delay, calibrating the vocalist's utterance time according to the delay difference of the singer's pronunciation; if greater than the preset threshold, determining that the singer himself delays, uttering the singer's pronunciation The delay difference is not calibrated.
- Step S400 if it is a system delay, calibrating the vocalist's pronunciation time according to the delay difference of the singer's pronunciation.
- the vocalist's pronunciation time is calibrated according to the calculated delay difference according to the calculated delay difference, that is, the singer's pronunciation time is advanced by the corresponding delay difference according to the calculated delay difference. time. For example, if the singer's pronunciation has a delay of 20 seconds, then the singer's pronunciation time is advanced by 20 seconds, keeping the singer's pronunciation in sync with the original singer's pronunciation.
- Step S500 comparing the pronunciation of the vocalist's calibration pronunciation time with the pronunciation of the original singer, and giving a score.
- the karaoke scoring device compares the pronunciation of the synchronized singer with the pronunciation of the original singer, for example, the timbre, melody, rhythm, etc. of the pronunciation, and finally the karaoke scoring device performs on the display screen according to the comparison result.
- the score is displayed. For example, show Xiao Ming: 9.5 points.
- the karaoke scoring method provided in this embodiment improves the false positive rate of the karaoke scoring device scoring software, enhances the user experience, and increases the selling point of the karaoke scoring device.
- FIG. 2 is a schematic flowchart of a second embodiment of a karaoke scoring method according to the present invention.
- the karaoke scoring method provided by the second embodiment further includes before the step S500. :
- Step S500A acquiring a pronunciation frequency of the singer and a pronunciation frequency of the original singer, the pronunciation frequency of the singer and the pronunciation frequency of the original singer are acquired while acquiring the pronunciation time; and the calibration pronunciation time of the singer is The corresponding pronunciation frequency is compared with the pronunciation frequency of the original singer, and if there is a difference, the pronunciation frequency of the singer is compensated.
- the karaoke scoring device obtains the pronunciation frequency of the singer when singing, compares the pronunciation frequency corresponding to the vocalist's calibrated pronunciation time and the frequency of the original singer's pronunciation frequency, and if there is a difference, the pronunciation frequency of the singer Make compensation. For example, the karaoke scoring device obtains the original speaker's pronunciation frequency as 2200 Hz (hertz), and the singer's calibrated synchronized pronunciation time corresponds to the total of the pronunciation frequency is 2000 Hz, and the singer's calibration pronunciation time corresponds to the synchronization. The pronunciation frequency needs to be compensated for 200Hz.
- the frequency of the vocalist's calibrated pronunciation time corresponding to the synchronized pronunciation time is compensated, and the effect can be obtained by keeping the vocalist's pronunciation frequency and the original singer's pronunciation frequency the same, thereby generating resonance and enhancing the user.
- FIG. 3 is a schematic flowchart of a third embodiment of a karaoke scoring method according to the present invention.
- the karaoke scoring method provided by the third embodiment, before the step S100 includes:
- Step S110 determining whether there is a vocalist's pronunciation, and if so, obtaining the vocalist's pronunciation time and pronunciation frequency, and obtaining the original singer's pronunciation time and pronunciation frequency.
- the karaoke scoring device collects continuous pronunciation signals. If the confirmation pronunciation signal is within the preset vocal frequency range, the vocalist pronounces, if any, the vocalist's pronunciation time and pronunciation frequency, and obtains the original singer's Pronunciation time and pronunciation frequency. If it is not within the preset vocal frequency range, filtering is performed, wherein the preset vocal frequency range is between 100 Hz and 1000 Hz.
- the card-based OK scoring method provided in this embodiment pre-identifies the vocalist's pronunciation, obtains the vocalist's pronunciation time and pronunciation frequency, and acquires the original singer's pronunciation time and pronunciation frequency, and filters out the influence of unnecessary noise signals.
- FIG. 4 is a schematic flowchart of the step S110 of the karaoke scoring method of the present invention.
- the step S110 specifically includes the following steps:
- Step S111 Collect continuous pronunciation signals.
- the microphone of the karaoke scoring device collects continuous pronunciation signals, wherein the pronunciation signal is a pronunciation signal with a certain frequency sent by the outside world, for example, 400 issued by the user when singing. Sound signal at Hz frequency.
- Step S112 converting the pronunciation signal into a digital signal.
- the karaoke scoring device performs analog-to-digital conversion of the continuous sounding signal, that is, converts the continuous sounding signal into a digital signal.
- Step S113 Acquire a number of pronunciations in a set time according to the converted digital signal.
- the karaoke scoring device uses the microphone sampling rate to change the converted digital signal to the number of times of pronunciation within a set time, for example, the currently collected pronunciation signal is 200.
- the frequency of Hz and the karaoke scoring device uses a microphone sampling rate of 40K Hz. If the set time is 10 seconds, the number of times of the received pronunciation signal is 200 times in 10 seconds.
- Step S114 Comparing the number of times of the pronunciation in the acquired set time with the preset number of times of pronunciation, if the number of times of the acquired set time is within the preset number of times of pronunciation, it is confirmed that The singer is singing, and if the number of pronunciations of the conversion is not within the preset number of times, it is confirmed that no singer is singing, and the pronunciation time and pronunciation frequency are no longer acquired.
- the karaoke scoring device compares the obtained number of pronunciations in the set time with the preset number of pronunciation times; if the number of pronunciations in the set time period is within the preset number of pronunciation times, the singer is confirmed While singing, if the number of pronunciations of the conversion is not within the preset number of times, it is confirmed that no singer is singing, and the pronunciation time and pronunciation frequency are no longer acquired.
- the karaoke scoring device uses a microphone sampling rate of 40K. Hz, the preset vocal frequency range is between 100Hz and 1000Hz. Then, within 10 seconds, it can be known from the calculation that if the number of times is within 40 to 400, it belongs to the range of normal sampling, that is, the karaoke scoring device presets.
- the number of pronunciations ranges from 40 to 400, and the pronunciation signal is 200.
- the frequency of Hz can be known by calculation, the number of sampling times in the 10 seconds is 200 times, within the preset number of pronunciation times, it is confirmed that the singer is singing; and for the pronunciation signal that is not in the preset number of pronunciation times , it is confirmed as a noise signal, and thus filtered to reduce the influence of the noise signal.
- FIG. 5 is a schematic flowchart of a fourth embodiment of a karaoke scoring method according to the present invention.
- the karaoke scoring method provided by the fourth embodiment includes the following steps:
- Step S300A comparing a delay difference of the singer's pronunciation with a preset time threshold, if it is less than the preset threshold, determining that the system is delayed, and calibrating the singer according to the delay difference of the singer's pronunciation.
- the pronunciation time if it is greater than the preset threshold, it is determined that the singer himself delays, and the delay difference of the singer's pronunciation is not calibrated.
- the karaoke scoring device compares the singer's pronunciation delay difference with a preset time threshold. If it is less than the preset threshold, it determines that it is a system delay, such as a delay caused by a device such as a microphone during transmission, then The time difference of the singer's pronunciation is corrected, and the vocalist's pronunciation time is calibrated. If it is greater than the preset threshold, it is considered to be caused by the singer's personal factor, such as the singer deliberately dragging, and the delay difference of the singer's pronunciation is not calibrated.
- the karaoke scoring method provided in this embodiment makes a fair evaluation of the vocalist's pronunciation, thereby reducing false positives in the score and improving the correct rate of the score.
- the embodiment further provides a karaoke scoring device, where the karaoke scoring device includes:
- the obtaining module 10 is configured to obtain a vocalist's pronunciation time and an original singer's pronunciation time
- the calculating module 20 is configured to calculate a delay difference of the singer's pronunciation according to the singer's pronunciation time and the original singer's pronunciation time;
- the determining module 30 is configured to determine whether the system delays or the singer's own delay according to the calculated delay difference of the singer's pronunciation;
- the calibration module 40 is configured to calibrate the vocalist's pronunciation time according to the delay difference of the singer's pronunciation if it is a system delay;
- the scoring module 50 is configured to compare the pronunciation of the vocalist's calibrated pronunciation time with the pronunciation of the original singer, and give a score.
- the acquiring module 10 of the karaoke scoring device detects the vocalist's pronunciation signal and the original singer's pronunciation signal, and respectively obtains the singer's pronunciation time and the original singer's pronunciation time, wherein the singer's pronunciation time is karaoke.
- the singer's voice played in the scoring device is not the time when the singer actually starts singing, because the singer will have a delay in the transmission of the microphone while singing, and the original singer's pronunciation time is the singer's song, on-demand The time when the original singer began to speak, for example, the singer on-demand the song "Love You for Ten Thousand Years" by Hong Kong and Taiwan star Andy Lau, when the singer picked up the microphone and started singing the lyrics for the singer's pronunciation time.
- the karaoke scoring device can be a smart TV, a DVD player, a computer, a smart phone, a tablet, and the like.
- the calculation module 20 of the karaoke scoring device calculates the delay difference of the singer's pronunciation according to the obtained vocalist's pronunciation time and the original singer's pronunciation time, and the delay difference is the original singer's pronunciation time minus
- the difference between the vocalist's pronunciation time and the singer's pronunciation time is not based on the actual time, but is counted from the start time of the Kari OK song, which is X points X.
- the format of the second for example, the vocalist's pronunciation time is 1 minute and 20 seconds, while the original singer's pronunciation time is 1 minute and 0 seconds, then the singer's pronunciation has a delay of 1 minute and 20 seconds minus 1 minute and 0 seconds.
- the result is 20 seconds, that is, the singer's pronunciation has a delay of 20 seconds.
- the judging module 30 of the karaoke scoring device first determines whether the system delays or the singer's own delay according to the calculated delay difference of the singer's pronunciation, for example, compares the singer's pronunciation delay difference with a preset time threshold. If it is less than the preset threshold, determining that it is a system delay, calibrating the vocalist's pronunciation time according to the singer's pronunciation delay difference; if greater than the preset threshold, determining that the singer is delayed by himself, The difference in delay of the singer's pronunciation is not calibrated.
- the vocalist's pronunciation time is calibrated according to the delay difference according to the calculated delay difference, and the vocalist's pronunciation time is advanced according to the calculated delay difference.
- the delay is time difference. For example, if the singer's pronunciation has a delay of 20 seconds, then the singer's pronunciation time is advanced by 20 seconds, keeping the singer's pronunciation in sync with the original singer's pronunciation.
- the scoring module 50 of the karaoke scoring device compares the pronunciation of the synchronized singer with the pronunciation of the original singer, for example, the timbre, melody, rhythm, etc. of the pronunciation, and finally the karaoke scoring device according to the comparison result,
- the score display is displayed on the display screen. For example, show Xiao Ming: 9.5 points.
- the karaoke scoring device provided in the embodiment improves the false positive rate of the karaoke scoring device scoring software, enhances the user experience, and increases the selling point of the karaoke scoring device.
- FIG. 7 is a schematic diagram of a functional module of a second embodiment of the karaoke scoring apparatus of the present invention.
- the karaoke scoring apparatus provided by the second embodiment further includes:
- a frequency compensation module 60 configured to acquire a pronunciation frequency of the singer and a pronunciation frequency of the original singer, the pronunciation frequency of the singer and the pronunciation frequency of the original singer are acquired while acquiring the pronunciation time; the singer The pronunciation frequency corresponding to the calibration pronunciation time is compared with the pronunciation frequency of the original singer, and if there is a difference, the pronunciation frequency of the singer is compensated.
- the frequency compensation module 60 of the karaoke scoring device obtains the pronunciation frequency of the singer singing, compares the pronunciation frequency corresponding to the vocalist's calibrated pronunciation time and the frequency of the original singer's pronunciation frequency, and if there is a difference, the The vocalist's pronunciation frequency is compensated.
- the karaoke scoring device obtains the original speaker's pronunciation frequency as 2200 Hz (hertz), and the singer's calibrated synchronized pronunciation time corresponds to the total of the pronunciation frequency is 2000 Hz, and the singer's calibration pronunciation time corresponds to the synchronization.
- the pronunciation frequency needs to be compensated for 200Hz.
- the frequency of the vocalist's calibrated pronunciation time corresponding to the synchronized pronunciation time is compensated, and the effect can be obtained by keeping the vocalist's pronunciation frequency and the original singer's pronunciation frequency the same, thereby generating resonance and enhancing the user.
- FIG. 8 is a schematic diagram of a function module of a third embodiment of the karaoke scoring apparatus of the present invention, and the karaoke scoring apparatus provided by the third embodiment, based on the first embodiment, the karaoke scoring apparatus Also includes:
- the preset module 10 is configured to determine whether there is a vocalist's pronunciation, and if so, obtain the vocalist's pronunciation time and pronunciation frequency, and obtain the original singer's pronunciation time and pronunciation frequency.
- the preset module 10 of the karaoke scoring device collects continuous pronunciation signals. If the confirmation pronunciation signal is within the preset vocal frequency range, the vocalist pronounces, if any, the vocalist's pronunciation time and pronunciation frequency, and simultaneously Get the pronunciation time and pronunciation frequency of the original singer. If it is not within the preset vocal frequency range, filtering is performed, wherein the preset vocal frequency range is between 100 Hz and 1000 Hz.
- the card OK calibration device pre-identifies the vocalist's pronunciation, obtains the vocalist's pronunciation time and pronunciation frequency, and acquires the original singer's pronunciation time and pronunciation frequency, and filters out the influence of unnecessary noise signals.
- FIG. 9 is a schematic diagram of functional modules of the preset module in FIG. 8 , where the preset module 10 includes:
- the collecting unit 11 is configured to collect continuous pronunciation signals
- a converting unit 12 configured to convert the pronunciation signal into a digital signal
- the obtaining unit 13 is configured to acquire the number of times of pronunciation in the set time according to the converted digital signal
- the comparing unit 14 is configured to compare the number of pronunciations in the acquired set time with a preset pronunciation sub-interval, if the number of pronunciations in the acquired set time is within a preset number of pronunciation times, Then, it is confirmed that the singer is singing, and if the number of times of the conversion is not within the preset number of times, it is confirmed that no singer is singing, and the pronunciation time and the pronunciation frequency are no longer acquired.
- the microphone acquisition unit 11 of the karaoke scoring device collects a continuous pronunciation signal, wherein the pronunciation signal is a pronunciation signal with a certain frequency sent by the outside world, for example, 400 issued by the user when singing. Sound signal at Hz frequency.
- the converting unit 12 of the karaoke scoring device performs analog-to-digital conversion of the continuous sounding signal, that is, converts the continuous sounding signal into a digital signal.
- the acquiring unit 13 of the karaoke scoring device uses the microphone sampling rate to change the converted digital signal to the number of times of the pronunciation within a set time, for example, the currently collected pronunciation signal is 200.
- the frequency of Hz and the karaoke scoring device uses a microphone sampling rate of 40K Hz. If the set time is 10 seconds, the number of times of the received pronunciation signal is 200 times in 10 seconds.
- the comparing unit 14 of the karaoke scoring device compares the acquired number of times of the set time with the preset number of times of pronunciation; if the number of times of the obtained set time is within the preset number of times of pronunciation, then It is confirmed that the singer is singing, and if the number of times of the conversion is not within the preset number of times, it is confirmed that no singer is singing, and the pronunciation time and pronunciation frequency are no longer acquired.
- the karaoke scoring device uses a microphone sampling rate of 40K. Hz, the preset vocal frequency range is between 100Hz and 1000Hz.
- the number of times is within 40 to 400, it belongs to the range of normal sampling, that is, the karaoke scoring device presets.
- the number of pronunciations ranges from 40 to 400, and the pronunciation signal is 200.
- the frequency of Hz can be known by calculation, the number of sampling times in the 10 seconds is 200 times, within the preset number of pronunciation times, it is confirmed that the singer is singing; and for the pronunciation signal that is not in the preset number of pronunciation times , it is confirmed as a noise signal, and thus filtered to reduce the influence of the noise signal.
- the karaoke scoring apparatus is configured to compare the delay difference of the singer's pronunciation with a preset time threshold, and if it is less than the preset threshold, Determining that it is a system delay, calibrating the vocalist's pronunciation time according to the delay difference of the singer's pronunciation; if it is greater than the preset threshold, determining that the singer is delayed by himself, and the delay difference of the singer's pronunciation is not performed. calibration.
- the judging module 3 of the karaoke scoring device compares the delay difference of the vocalist's pronunciation with a preset time threshold, and if it is less than the preset threshold, it determines that it is a system delay, such as a delay caused by a device such as a microphone during transmission. Then, the vocalist's pronunciation time is calibrated according to the delay difference of the singer's pronunciation. If it is greater than the preset threshold, it is considered to be caused by the singer's personal factor, such as the singer deliberately dragging, and the delay difference of the singer's pronunciation is not calibrated.
- the karaoke scoring device provided in this embodiment makes a fair evaluation of the vocalist's pronunciation, thereby reducing false positives in the score and improving the correct rate of the score.
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Reverberation, Karaoke And Other Acoustics (AREA)
Abstract
一种卡拉OK评分方法，通过获取歌唱者的发音时间和原唱者的发音时间（S100）；根据所述歌唱者的发音时间和所述原唱者的发音时间，计算歌唱者发音的时延差（S200）；根据计算出来的歌唱者发音的时延差，判断是***延迟还是歌唱者自己延迟（S300）；如果是***延迟则根据所述歌唱者发音的时延差，校准所述歌唱者的发音时间（S400）；将所述歌唱者的校准发音时间的发音与所述原唱者的发音进行比较，给出评分（S500）。还提供一种卡拉OK评分装置，其改善卡拉OK评分装置评分软件的误判率、增强用户体验、增加卡拉OK评分装置的卖点。
Description
技术领域
本发明涉及媒体播放领域，尤其涉及发明卡拉OK评分方法和装置。
背景技术
现代人越来越重视生活娱乐，工作之余常常参与各种娱乐活动，从而导致KTV卡拉OK越来越流行，KTV的流行又让卡拉OK评分装置（比如智能电视）受到用户的喜爱和青睐，销量节节攀高，但是，歌唱者在使用麦克风唱KTV歌曲时，由于发音在传送过程中延迟造成声音的延迟，而歌唱者的声音延迟造成卡拉OK评分装置评分软件的误判，引起评分错误，比如说，歌唱者开启天籁K歌应用软件，选择了港台明星刘德华的歌曲《爱你一万年》，当歌唱者拿起话筒，对着歌词开始演唱时，由于卡拉OK评分装置评分***的错误，卡拉OK评分装置的屏幕上显示的评分却为0分，造成歌唱者的困惑。如果解决歌唱者的发音在传送过程中的延迟而造成卡拉OK评分装置评分软件的误判是一个亟待解决的问题。
发明内容
本发明的主要目的在于解决歌唱者的发音在传送过程中的延迟而造成卡拉OK评分装置评分软件的误判的问题。
本发明提供的一种卡拉OK评分方法，所述卡拉OK评分方法包括以下步骤：
获取歌唱者的发音时间和原唱者的发音时间；
根据所述歌唱者的发音时间和所述原唱者的发音时间，计算歌唱者发音的时延差；
根据计算出来的歌唱者发音的时延差，判断是***延迟还是歌唱者自己延迟；
如果是***延迟则根据所述歌唱者发音的时延差，校准所述歌唱者的发音时间；
将所述歌唱者的校准发音时间的发音和所述原唱者的发音进行比较，给出评分。
优选地，所述将所述歌唱者的校准发音时间的发音和所述原唱者的发音进行对比，给出评分的步骤之前还包括：
获取所述歌唱者的发音频率和原唱者的发音频率，所述歌唱者的发音频率和原唱者的发音频率在获取发音时间的同时获取；将所述歌唱者的校准发音时间对应的发音频率与所述原唱者的发音频率进行比较，如存在差异，则对所述歌唱者的发音频率进行补偿。
优选地，所述获取歌唱者的发音时间和原唱者的发音时间的步骤之前还包括：
判断是否有歌唱者发音，如果有，则获取歌唱者的发音时间和发音频率，同时获取原唱者的发音时间和发音频率。
优选地，所述判断是否有歌唱者发音，如果有，则获取发音时间和发音频率，同时获取原唱者的发音时间和发音频率的步骤具体包括：
采集连续的发音信号；
将所述发音信号转换为数字信号；
根据所述转换的数字信号，获取设定时间内的发音次数；
将所述获取的设定时间内的发音次数与预设的发音次数区间进行比较，如果所述获取的设定时间内的发音次数在预设的发音次数区间之内，则确认有歌唱者在唱歌，如果所述转换的发音次数不在预设的次数区间之内，则确认没有歌唱者在唱歌，不再获取发音时间和发音频率。
优选地，所述根据计算出来的歌唱者发音的时延差，判断是***延迟还是歌唱者自己延迟的步骤包括：
将所述歌唱者发音的时延差和预设的时间阈值进行比较，如果小于所述预设阈值，则确定是***延迟，根据所述歌唱者发音的时延差，校准歌唱者的发音时间；如果大于所述预设阈值，则确定是歌唱者自己延迟，对歌唱者发音的时延差不进行校准。
为了解决上述的技术问题，本发明进一步提供一种卡拉OK评分装置，所述卡拉OK评分装置包括：
获取模块，用于获取歌唱者的发音时间和原唱者的发音时间；
计算模块，用于根据所述歌唱者的发音时间和所述原唱者的发音时间，计算所述歌唱者发音的时延差；
判断模块，用于根据计算出来的歌唱者发音的时延差，判断是***延迟还是歌唱者自己延迟；
校准模块，用于如果是***延迟则根据所述歌唱者发音的时延差，校准所述歌唱者的发音时间；
评分模块，用于将所述歌唱者的校准发音时间的发音和所述原唱者的发音进行比较，给出评分。
优选地，所述卡拉OK评分装置还包括：
频率补偿模块，用于获取所述歌唱者的发音频率和原唱者的发音频率，所述歌唱者的发音频率和原唱者的发音频率在获取发音时间的同时获取；将所述歌唱者的校准发音时间对应的发音频率与所述原唱者的发音频率进行比较，如存在差异，则对所述歌唱者的发音频率进行补偿。
优选地，所述卡拉OK评分装置还包括：
预设模块，用于判断是否有歌唱者发音，如果有，则获取歌唱者的发音时间和发音频率，同时获取原唱者的发音时间和发音频率。
优选地，所述预设模块包括：
采集单元，用于采集连续的发音信号；
转换单元，用于将所述发音信号转换为数字信号；
获取单元，用于根据所述转换的数字信号，获取设定时间内的发音次数；
比较单元，用于将所述获取的设定时间内的发音次数与预设的发音次区间进行比较，如果所述获取的设定时间内的发音次数在预设的发音次数区间之内，则确认有歌唱者在唱歌，如果所述转换的发音次数不在预设的次数区间之内，则确认没有歌唱者在唱歌，不再获取发音时间和发音频率。
优选地，所述判断模块，用于将所述歌唱者发音的时延差和预设的时间阈值进行比较，如果小于所述预设阈值，则确定是***延迟，根据所述歌唱者发音的时延差，校准歌唱者的发音时间；如果大于所述预设阈值，则确定是歌唱者自己延迟，对歌唱者发音的时延差不进行校准。
本发明提供的卡拉OK评分方法，通过获取歌唱者的发音时间和原唱者的发音时间；根据所述歌唱者的发音时间和所述原唱者的发音时间，计算歌唱者发音的时延差；根据计算出来的歌唱者发音的时延差，判断是***延迟还是歌唱者自己延迟；如果是***延迟则根据所述歌唱者发音的时延差，校准所述歌唱者的发音时间；将所述歌唱者的校准发音时间的发音和所述原唱者的发音进行比较，给出评分。本发明改善卡拉OK评分装置评分软件的误判率、增强用户体验、增加卡拉OK评分装置的卖点。
附图说明
图1为本发明卡拉OK评分方法第一实施例的流程示意图；
图2为本发明卡拉OK评分方法第二实施例的流程示意图；
图3为本发明卡拉OK评分方法第三实施例的流程示意图；
图4为图3中所述判断是否有歌唱者发音，如果有，则获取歌唱者的发音时间和发音频率，同时获取原唱者的发音时间和发音频率的步骤的细化流程示意图；
图5为本发明卡拉OK评分方法第四实施例的流程示意图；
图6为本发明卡拉OK评分装置第一实施例的功能模块示意图；
图7为本发明卡拉OK评分装置第二实施例的功能模块示意图；
图8为本发明卡拉OK评分装置第三实施例的功能模块示意图；
图9为图8中所述预设模块的功能模块示意图。
本发明目的的实现、功能特点及优点将结合实施例，参照附图做进一步说明。
具体实施方式
应当理解，此处所描述的具体实施例仅仅用以解释本发明，并不用于限定本发明。
本发明提供一种卡拉OK评分方法，参照图1，在第一实施例中，所述卡拉OK评分方法包括以下步骤：
步骤S100、获取歌唱者的发音时间和原唱者的发音时间。
卡拉OK评分装置侦测到歌唱者的发音信号和原唱者的发音信号后，分别获取歌唱者的发音时间和原唱者的发音时间，其中，歌唱者的发音时间为卡拉OK评分装置中播放传出来的歌唱者的声音，并非歌唱者实际开始唱歌的时间，因为歌唱者在演唱时会有话筒的传输延时，而原唱者的发音时间为歌唱者点歌时，点播的歌曲中原唱者开始发声的时间，比如说歌唱者点播了港台明星刘德华的歌曲《爱你一万年》，当歌唱者拿起话筒，对着歌词开始演唱时为歌唱者的发音时间，而刘德华的歌曲《爱你一万年》中，卡拉OK评分装置开始播放刘德华歌曲的时间为原唱者的发音时间。其中，卡拉OK评分装置可以为智能电视、DVD播放机、电脑、智能手机、平板电脑等。
步骤S200、根据所述歌唱者的发音时间和所述原唱者的发音时间，计算所述歌唱者发音的时延差。
卡拉OK评分装置根据获取的歌唱者的发音时间和原唱者的发音时间，计算出所述歌唱者发音的时延差，所述时延差为原唱者的发音时间减去歌唱者的发音时间的差值，歌唱者的发音时间不是以实际的时间作为参考，而是从卡里OK歌曲的开始时间进行计时，为X分X
秒的格式，比如说，歌唱者的发音时间为1分20秒，而原唱者的发音时间为1分0秒，那么歌唱者发音的时延差为1分20秒减去1分0秒，得到的结果为20秒，即歌唱者发音的时延差为20秒。
步骤S300、根据计算出来的歌唱者发音的时延差，判断是***延迟还是歌唱者自己延迟。
卡拉OK评分装置根据计算出来的歌唱者发音的时延差，初步判断是***延迟还是歌唱者自己延迟，例如将所述歌唱者发音的时延差和预设的时间阈值进行比较，如果小于所述预设阈值，则确定是***延迟，根据所述歌唱者发音的时延差，校准歌唱者的发音时间；如果大于所述预设阈值，则确定是歌唱者自己延迟，对歌唱者发音的时延差不进行校准。
步骤S400、如果是***延迟则根据所述歌唱者发音的时延差，校准所述歌唱者的发音时间。
卡拉OK评分装置如果判断是***延迟，则根据计算出来的时延差，按照时延差校准歌唱者的发音时间，即按照计算出来的时延差，将歌唱者的发音时间提前相应时延差时间。比如说，歌唱者发音的时延差为20秒，那么将歌唱者的发音时间相应提前20秒，保持歌唱者的发音与原唱者的发音同步。
步骤S500、将所述歌唱者的校准发音时间的发音和所述原唱者的发音进行比较，给出评分。
卡拉OK评分装置将同步后的歌唱者的发音与原唱者的发音进行比较，比如说发音的音色、旋律、节奏等进行对较，最后卡拉OK评分装置根据比较的结果，在显示屏幕上进行得分显示。比如说显示小明：9.5分。
本实施例中提供的卡拉OK评分方法，改善卡拉OK评分装置评分软件的误判率、增强用户体验、增加卡拉OK评分装置的卖点。
如图2所述，图2为本发明卡拉OK评分方法第二实施例的流程示意图，在第一实施例的基础上，第二实施例提供的卡拉OK评分方法，所述步骤S500之前还包括：
步骤S500A、获取所述歌唱者的发音频率和原唱者的发音频率，所述歌唱者的发音频率和原唱者的发音频率在获取发音时间的同时获取；将所述歌唱者的校准发音时间对应的发音频率与所述原唱者的发音频率进行比较，如存在差异，则对所述歌唱者的发音频率进行补偿。
卡拉OK评分装置获取歌唱者歌唱时的发音频率，与歌唱者的校准发音时间对应的发音频率和原唱者的发音频率的频率总和进行比较，如存在差异，则对所述歌唱者的发音频率进行补偿。比如说，卡拉OK评分装置获取原唱者的发音频率总和为2200Hz（赫兹），而歌唱者的校准后的同步发音时间对应的发音频率的总和为2000Hz，则歌唱者的校准发音时间对应的同步发音频率需要补偿200Hz。
本实施例对歌唱者的校准后的同步发音时间对应的发音频率的进行频率补偿，所能取得的效果为让歌唱者的发音频率和原唱者的发音频率保持相同，从而产生共鸣，增强用户体验、增加卡拉OK评分装置的卖点。
如图3所述，图3为本发明卡拉OK评分方法第三实施例的流程示意图，在第一实施例的基础上，第三实施例提供的卡拉OK评分方法，所述步骤S100之前包括：
步骤S110、判断是否有歌唱者发音，如果有，则获取歌唱者的发音时间和发音频率，同时获取原唱者的发音时间和发音频率。
卡拉OK评分装置采集连续的发音信号，如确认发音信号在预设的人声频率范围内，则有歌唱者发音，如果有，则获取歌唱者的发音时间和发音频率，同时获取原唱者的发音时间和发音频率。如不在预设的人声频率范围内，则进行滤除，其中，预设的人声频率范围为100Hz至1000Hz之间。
本实施例提供的卡接OK评分方法，预先识别歌唱者发音，获取歌唱者的发音时间和发音频率，同时获取原唱者的发音时间和发音频率，滤除不必要的噪音信号的影响。
如图4所述，图4为本发明卡拉OK评分方法所述步骤S110的细化流程示意图，所述步骤S110具体包括以下步骤：
步骤S111、采集连续的发音信号。
卡拉OK评分装置的话筒采集连续的发音信号，其中，发音信号为外界发出的具有一定频率的发音信号，比如：用户唱歌时发出的400
Hz频率的声音信号。
步骤S112、将所述发音信号转换为数字信号。
卡拉OK评分装置将连续的发音信号进行模数转换，即将连续的发音信号转换为数字信号。
步骤S113、根据所述转换的数字信号，获取设定时间内的发音次数。
卡拉OK评分装置利用话筒采样率，将转换的数字信号在设定的时间内变为发音次数，比如说当前采集的发音信号为200
Hz的频率，而卡拉OK评分装置利用话筒采样率为40K Hz，如果设定时间为10秒，那么在10秒时间内，采集的发音信号的次数为200次。
步骤S114、将所述获取的设定时间内的发音次数与预设的发音次数区间进行比较，如果所述获取的设定时间内的发音次数在预设的发音次数区间之内，则确认有歌唱者在唱歌，如果所述转换的发音次数不在预设的次数区间之内，则确认没有歌唱者在唱歌，不再获取发音时间和发音频率。
卡拉OK评分装置将获取到的设定时间内的发音次数与预设的发音次数区间进行比较；如果获取的设定时间内的发音次数在预设的发音次数区间之内，则确认有歌唱者在唱歌，如果所述转换的发音次数不在预设的次数区间之内，则确认没有歌唱者在唱歌，不再获取发音时间和发音频率。比如说卡拉OK评分装置利用话筒采样率为40K
Hz，预设的人声频率范围为100Hz至1000Hz之间，那么在10秒时间内，通过计算可知，如果次数在40至400之内都属于正常取样的范围区间，即卡拉OK评分装置预设的发音次数区间为40至400，对于发音信号为200
Hz的频率，通过计算可知，在10秒时间内的采样次数为200次，在预设的发音次数区间之内，则确认有歌唱者在唱歌；而对于不在预设的发音次数区间的发音信号，则确认为噪音信号，从而进行滤除，减少噪音信号的影响。
如图5所述，图5为本发明卡拉OK评分方法第四实施例的流程示意图，第四实施例提供的卡拉OK评分方法，所述步骤S300包括以下步骤：
步骤S300A、将所述歌唱者发音的时延差和预设的时间阈值进行比较，如果小于所述预设阈值，则确定是***延迟，根据所述歌唱者发音的时延差，校准歌唱者的发音时间；如果大于所述预设阈值，则确定是歌唱者自己延迟，对歌唱者发音的时延差不进行校准。
卡拉OK评分装置将歌唱者发音的时延差和预设的时间阈值进行比较，如果小于预设阈值，则确定是***延迟，例如由话筒等设备在传输过程中引起的延时，那么根据所述歌唱者发音的时延差，校准歌唱者的发音时间。如果大于所述预设阈值，则认为是由歌唱者个人因素引起，例如歌唱者故意拖音，对歌唱者发音的时延差不进行校准。
本实施例提供的卡拉OK评分方法，对歌唱者的发音做出公正的评判，从而减小评分中的误判，提高评分的正确率。
进一步参见图6，本实施例还提供一种卡拉OK评分装置，所述卡拉OK评分装置包括：
获取模块10，用于获取歌唱者的发音时间和原唱者的发音时间；
计算模块20，用于根据所述歌唱者的发音时间和所述原唱者的发音时间，计算所述歌唱者发音的时延差；
判断模块30，用于根据计算出来的歌唱者发音的时延差，判断是***延迟还是歌唱者自己延迟；
校准模块40，用于如果是***延迟则根据所述歌唱者发音的时延差，校准所述歌唱者的发音时间；
评分模块50，用于将所述歌唱者的校准发音时间的发音和所述原唱者的发音进行比较，给出评分。
卡拉OK评分装置的获取模块10侦测到歌唱者的发音信号和原唱者的发音信号后，分别获取歌唱者的发音时间和原唱者的发音时间，其中，歌唱者的发音时间为卡拉OK评分装置中播放传出来的歌唱者的声音，并非歌唱者实际开始唱歌的时间，因为歌唱者在演唱时会有话筒的传输延时，而原唱者的发音时间为歌唱者点歌时，点播的歌曲中原唱者开始发声的时间，比如说歌唱者点播了港台明星刘德华的歌曲《爱你一万年》，当歌唱者拿起话筒，对着歌词开始演唱时为歌唱者的发音时间，而刘德华的歌曲《爱你一万年》中卡拉OK评分装置开始播放刘德华歌曲的时间为原唱者的发音时间。其中，卡拉OK评分装置可以为智能电视、DVD播放机、电脑、智能手机、平板电脑等。
卡拉OK评分装置的计算模块20根据获取的歌唱者的发音时间和原唱者的发音时间，计算出所述歌唱者发音的时延差，所述时延差为原唱者的发音时间减去歌唱者的发音时间的差值，歌唱者的发音时间不是以实际的时间作为参考，而是从卡里OK歌曲的开始时间进行计时，为X分X
秒的格式，比如说，歌唱者的发音时间为1分20秒，而原唱者的发音时间为1分0秒，那么歌唱者发音的时延差为1分20秒减去1分0秒，得到的结果为20秒，即歌唱者发音的时延差为20秒。
卡拉OK评分装置的判断模块30根据计算出来的歌唱者发音的时延差，初步判断是***延迟还是歌唱者自己延迟，例如将所述歌唱者发音的时延差和预设的时间阈值进行比较，如果小于所述预设阈值，则确定是***延迟，根据所述歌唱者发音的时延差，校准歌唱者的发音时间；如果大于所述预设阈值，则确定是歌唱者自己延迟，对歌唱者发音的时延差不进行校准。
卡拉OK评分装置的校准模块50如果判断是***延迟，则根据计算出来的时延差，按照时延差校准歌唱者的发音时间，按照计算出来的时延差，将歌唱者的发音时间提前相应时延差时间。比如说，歌唱者发音的时延差为20秒，那么将歌唱者的发音时间相应提前20秒，保持歌唱者的发音与原唱者的发音同步。
卡拉OK评分装置的评分模块50将同步后的歌唱者的发音与原唱者的发音进行比较，比如说发音的音色、旋律、节奏等进行对较，最后卡拉OK评分装置根据比较的结果，在显示屏幕上进行得分显示。比如说显示小明：9.5分。
本实施例中提供的卡拉OK评分装置，改善卡拉OK评分装置评分软件的误判率、增强用户体验、增加卡拉OK评分装置的卖点。
如图7所示，图7为本发明卡拉OK评分装置第二实施例的功能模块示意图，在一实施例的基础上，第二实施例提供的卡拉OK评分装置，还包括：
频率补偿模块60，用于获取所述歌唱者的发音频率和原唱者的发音频率，所述歌唱者的发音频率和原唱者的发音频率在获取发音时间的同时获取；将所述歌唱者的校准发音时间对应的发音频率与所述原唱者的发音频率进行比较，如存在差异，则对所述歌唱者的发音频率进行补偿。
卡拉OK评分装置的频率补偿模块60获取歌唱者歌唱时的发音频率，与歌唱者的校准发音时间对应的发音频率和原唱者的发音频率的频率总和进行比较，如存在差异，则对所述歌唱者的发音频率进行补偿。比如说，卡拉OK评分装置获取原唱者的发音频率总和为2200Hz（赫兹），而歌唱者的校准后的同步发音时间对应的发音频率的总和为2000Hz，则歌唱者的校准发音时间对应的同步发音频率需要补偿200Hz。
本实施例对歌唱者的校准后的同步发音时间对应的发音频率的进行频率补偿，所能取得的效果为让歌唱者的发音频率和原唱者的发音频率保持相同，从而产生共鸣，增强用户体验、增加卡拉OK评分装置的卖点。
如图8所示，图8为本发明卡拉OK评分装置第三实施例的功能模块示意图，第三实施例提供的卡拉OK评分装置，在第一实施例的基础上，所述卡拉OK评分装置还包括：
预设模块10，用于判断是否有歌唱者发音，如果有，则获取歌唱者的发音时间和发音频率，同时获取原唱者的发音时间和发音频率。
卡拉OK评分装置的预设模块10采集连续的发音信号，如确认发音信号在预设的人声频率范围内，则有歌唱者发音，如果有，则获取歌唱者的发音时间和发音频率，同时获取原唱者的发音时间和发音频率。如不在预设的人声频率范围内，则进行滤除，其中，预设的人声频率范围为100Hz至1000Hz之间。
本实施例提供的卡接OK评分装置，预先识别歌唱者发音，获取歌唱者的发音时间和发音频率，同时获取原唱者的发音时间和发音频率，滤除不必要的噪音信号的影响。
如图9所示，图9为图8中所述预设模块的功能模块示意图，所述预设模块10包括：
采集单元11，用于采集连续的发音信号；
转换单元12，用于将所述发音信号转换为数字信号；
获取单元13，用于根据所述转换的数字信号，获取设定时间内的发音次数；
比较单元14，用于将所述获取的设定时间内的发音次数与预设的发音次区间进行比较，如果所述获取的设定时间内的发音次数在预设的发音次数区间之内，则确认有歌唱者在唱歌，如果所述转换的发音次数不在预设的次数区间之内，则确认没有歌唱者在唱歌，不再获取发音时间和发音频率。
卡拉OK评分装置的话筒采集单元11采集连续的发音信号，其中，发音信号为外界发出的具有一定频率的发音信号，比如：用户唱歌时发出的400
Hz频率的声音信号。
卡拉OK评分装置的转换单元12将连续的发音信号进行模数转换，即将连续的发音信号转换为数字信号。
卡拉OK评分装置的获取单元13利用话筒采样率，将转换的数字信号在设定的时间内变为发音次数，比如说当前采集的发音信号为200
Hz的频率，而卡拉OK评分装置利用话筒采样率为40K Hz，如果设定时间为10秒，那么在10秒时间内，采集的发音信号的次数为200次。
卡拉OK评分装置的比较单元14将获取到的设定时间内的发音次数与预设的发音次数区间进行比较；如果获取的设定时间内的发音次数在预设的发音次数区间之内，则确认有歌唱者在唱歌，如果所述转换的发音次数不在预设的次数区间之内，则确认没有歌唱者在唱歌，不再获取发音时间和发音频率。比如说卡拉OK评分装置利用话筒采样率为40K
Hz，预设的人声频率范围为100Hz至1000Hz之间，那么在10秒时间内，通过计算可知，如果次数在40至400之内都属于正常取样的范围区间，即卡拉OK评分装置预设的发音次数区间为40至400，对于发音信号为200
Hz的频率，通过计算可知，在10秒时间内的采样次数为200次，在预设的发音次数区间之内，则确认有歌唱者在唱歌；而对于不在预设的发音次数区间的发音信号，则确认为噪音信号，从而进行滤除，减少噪音信号的影响。
进一步参见图6，本实施例提供的卡拉OK评分装置，所述判断模块30，用于将所述歌唱者发音的时延差和预设的时间阈值进行比较，如果小于所述预设阈值，则确定是***延迟，根据所述歌唱者发音的时延差，校准歌唱者的发音时间；如果大于所述预设阈值，则确定是歌唱者自己延迟，对歌唱者发音的时延差不进行校准。
卡拉OK评分装置的判断模块3将歌唱者发音的时延差和预设的时间阈值进行比较，如果小于预设阈值，则确定是***延迟，例如由话筒等设备在传输过程中引起的延时，那么根据所述歌唱者发音的时延差，校准歌唱者的发音时间。如果大于所述预设阈值，则认为是由歌唱者个人因素引起，例如歌唱者故意拖音，对歌唱者发音的时延差不进行校准。
本实施例提供的卡拉OK评分装置，对歌唱者的发音做出公正的评判，从而减小评分中的误判，提高评分的正确率。
以上仅为本发明的优选实施例，并非因此限制本发明的专利范围，凡是利用本发明说明书及附图内容所作的等效结构或等效流程变换，或直接或间接运用在其他相关的技术领域，均同理包括在本发明的专利保护范围内。
Claims (20)
- 一种卡拉OK评分方法，其特征在于，所述卡拉OK评分方法包括以下步骤：获取歌唱者的发音时间和原唱者的发音时间；根据所述歌唱者的发音时间和所述原唱者的发音时间，计算歌唱者发音的时延差；根据计算出来的歌唱者发音的时延差，判断是***延迟还是歌唱者自己延迟；如果是***延迟则根据所述歌唱者发音的时延差，校准所述歌唱者的发音时间；将所述歌唱者的校准发音时间的发音和所述原唱者的发音进行比较，给出评分。
- 如权利要求1所述的卡拉OK评分方法，其特征在于，所述将将所述歌唱者的校准发音时间的发音和所述原唱者的发音进行比较，给出评分的步骤之前还包括：获取歌唱者的发音频率和原唱者的发音频率，所述歌唱者的发音频率和原唱者的发音频率在获取发音时间的同时获取；将所述歌唱者的校准发音时间对应的发音频率与所述原唱者的发音频率进行比较，如存在差异，则对所述歌唱者的发音频率进行补偿。
- 如权利要求1所述的卡拉OK评分方法，其特征在于，所述获取歌唱者的发音时间和原唱者的发音时间的步骤之前还包括：判断是否有歌唱者发音，如果有，则获取歌唱者的发音时间和发音频率，同时获取原唱者的发音时间和发音频率。
- 如权利要求2所述的卡拉OK评分方法，其特征在于，所述获取歌唱者的发音时间和原唱者的发音时间的步骤之前还包括：判断是否有歌唱者发音，如果有，则获取歌唱者的发音时间和发音频率，同时获取原唱者的发音时间和发音频率。
- 如权利要求2所述的卡拉OK评分方法，其特征在于，所述判断是否有歌唱者发音，如果有，则获取发音时间和发音频率，同时获取原唱者的发音时间和发音频率的步骤具体包括：采集连续的发音信号；将所述发音信号转换为数字信号；根据所述转换的数字信号，获取设定时间内的发音次数；将所述获取的设定时间内的发音次数与预设的发音次数区间进行比较，如果所述获取的设定时间内的发音次数在预设的发音次数区间之内，则确认有歌唱者在唱歌，如果所述转换的发音次数不在预设的次数区间之内，则确认没有歌唱者在唱歌，不再获取发音时间和发音频率。
- 如权利要求3所述的卡拉OK评分方法，其特征在于，所述判断是否有歌唱者发音，如果有，则获取发音时间和发音频率，同时获取原唱者的发音时间和发音频率的步骤具体包括：采集连续的发音信号；将所述发音信号转换为数字信号；根据所述转换的数字信号，获取设定时间内的发音次数；将所述获取的设定时间内的发音次数与预设的发音次数区间进行比较，如果所述获取的设定时间内的发音次数在预设的发音次数区间之内，则确认有歌唱者在唱歌，如果所述转换的发音次数不在预设的次数区间之内，则确认没有歌唱者在唱歌，不再获取发音时间和发音频率。
- 如权利要求4所述的卡拉OK评分方法，其特征在于，所述判断是否有歌唱者发音，如果有，则获取发音时间和发音频率，同时获取原唱者的发音时间和发音频率的步骤具体包括：采集连续的发音信号；将所述发音信号转换为数字信号；根据所述转换的数字信号，获取设定时间内的发音次数；将所述获取的设定时间内的发音次数与预设的发音次数区间进行比较，如果所述获取的设定时间内的发音次数在预设的发音次数区间之内，则确认有歌唱者在唱歌，如果所述转换的发音次数不在预设的次数区间之内，则确认没有歌唱者在唱歌，不再获取发音时间和发音频率。
- 如权利要求5所述的卡拉OK评分方法，其特征在于，所述根据计算出来的歌唱者发音的时延差，判断是***延迟还是歌唱者自己延迟的步骤包括：将所述歌唱者发音的时延差和预设的时间阈值进行比较，如果小于所述预设阈值，则确定是***延迟，根据所述歌唱者发音的时延差，校准歌唱者的发音时间；如果大于所述预设阈值，则确定是歌唱者自己延迟，对歌唱者发音的时延差不进行校准。
- 如权利要求6所述的卡拉OK评分方法，其特征在于，所述根据计算出来的歌唱者发音的时延差，判断是***延迟还是歌唱者自己延迟的步骤包括：将所述歌唱者发音的时延差和预设的时间阈值进行比较，如果小于所述预设阈值，则确定是***延迟，根据所述歌唱者发音的时延差，校准歌唱者的发音时间；如果大于所述预设阈值，则确定是歌唱者自己延迟，对歌唱者发音的时延差不进行校准。
- 如权利要求7所述的卡拉OK评分方法，其特征在于，所述根据计算出来的歌唱者发音的时延差，判断是***延迟还是歌唱者自己延迟的步骤包括：将所述歌唱者发音的时延差和预设的时间阈值进行比较，如果小于所述预设阈值，则确定是***延迟，根据所述歌唱者发音的时延差，校准歌唱者的发音时间；如果大于所述预设阈值，则确定是歌唱者自己延迟，对歌唱者发音的时延差不进行校准。
- 一种卡拉OK评分装置，其特征在于，所述卡拉OK评分装置包括：获取模块，用于获取歌唱者的发音时间和原唱者的发音时间；计算模块，用于根据所述歌唱者的发音时间和所述原唱者的发音时间，计算所述歌唱者发音的时延差；判断模块，用于根据计算出来的歌唱者发音的时延差，判断是***延迟还是歌唱者自己延迟；校准模块，用于如果是***延迟则根据所述歌唱者发音的时延差，校准所述歌唱者的发音时间；评分模块，用于将所述歌唱者的校准发音时间的发音和所述原唱者的发音进行比较，给出评分。
- 如权利要求11所述的卡拉OK评分装置，其特征在于，所述卡拉OK评分装置还包括：频率补偿模块，用于获取歌唱者的发音频率和原唱者的发音频率，所述歌唱者的发音频率和原唱者的发音频率在获取发音时间的同时获取；将所述歌唱者的校准发音时间对应的发音频率与所述原唱者的发音频率进行比较，如存在差异，则对所述歌唱者的发音频率进行补偿。
- 如权利要求11所述的卡拉OK评分装置，其特征在于，所述卡拉OK评分装置还包括：预设模块，用于判断是否有歌唱者发音，如果有，则获取歌唱者的发音时间和发音频率，同时获取原唱者的发音时间和发音频率。
- 如权利要求12所述的卡拉OK评分装置，其特征在于，所述卡拉OK评分装置还包括：预设模块，用于判断是否有歌唱者发音，如果有，则获取歌唱者的发音时间和发音频率，同时获取原唱者的发音时间和发音频率。
- 如权利要求12所述的卡拉OK评分装置，其特征在于，所述预设模块包括：采集单元，用于采集连续的发音信号；转换单元，用于将所述发音信号转换为数字信号；获取单元，用于根据所述转换的数字信号，获取设定时间内的发音次数；比较单元，用于将所述获取的设定时间内的发音次数与预设的发音次区间进行比较，如果所述获取的设定时间内的发音次数在预设的发音次数区间之内，则确认有歌唱者在唱歌，如果所述转换的发音次数不在预设的次数区间之内，则确认没有歌唱者在唱歌，不再获取发音时间和发音频率。
- 如权利要求13所述的卡拉OK评分装置，其特征在于，所述预设模块包括：采集单元，用于采集连续的发音信号；转换单元，用于将所述发音信号转换为数字信号；获取单元，用于根据所述转换的数字信号，获取设定时间内的发音次数；比较单元，用于将所述获取的设定时间内的发音次数与预设的发音次区间进行比较，如果所述获取的设定时间内的发音次数在预设的发音次数区间之内，则确认有歌唱者在唱歌，如果所述转换的发音次数不在预设的次数区间之内，则确认没有歌唱者在唱歌，不再获取发音时间和发音频率。
- 如权利要求14所述的卡拉OK评分装置，其特征在于，所述预设模块包括：采集单元，用于采集连续的发音信号；转换单元，用于将所述发音信号转换为数字信号；获取单元，用于根据所述转换的数字信号，获取设定时间内的发音次数；比较单元，用于将所述获取的设定时间内的发音次数与预设的发音次区间进行比较，如果所述获取的设定时间内的发音次数在预设的发音次数区间之内，则确认有歌唱者在唱歌，如果所述转换的发音次数不在预设的次数区间之内，则确认没有歌唱者在唱歌，不再获取发音时间和发音频率。
- 如权利要求15所述的卡拉OK评分装置，其特征在于，所述判断模块，用于将所述歌唱者发音的时延差和预设的时间阈值进行比较，如果小于所述预设阈值，则确定是***延迟，根据所述歌唱者发音的时延差，校准歌唱者的发音时间；如果大于所述预设阈值，则确定是歌唱者自己延迟，对歌唱者发音的时延差不进行校准。
- 如权利要求16所述的卡拉OK评分装置，其特征在于，所述判断模块，用于将所述歌唱者发音的时延差和预设的时间阈值进行比较，如果小于所述预设阈值，则确定是***延迟，根据所述歌唱者发音的时延差，校准歌唱者的发音时间；如果大于所述预设阈值，则确定是歌唱者自己延迟，对歌唱者发音的时延差不进行校准。
- 如权利要求17所述的卡拉OK评分装置，其特征在于，所述判断模块，用于将所述歌唱者发音的时延差和预设的时间阈值进行比较，如果小于所述预设阈值，则确定是***延迟，根据所述歌唱者发音的时延差，校准歌唱者的发音时间；如果大于所述预设阈值，则确定是歌唱者自己延迟，对歌唱者发音的时延差不进行校准。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410778465.X | 2014-12-15 | ||
| CN201410778465.XA CN105788581B (zh) | 2014-12-15 | 2014-12-15 | 卡拉ok评分方法和装置 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2016095280A1 true WO2016095280A1 (zh) | 2016-06-23 |
Family
ID=56125712
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2014/095645 WO2016095280A1 (zh) | 2014-12-15 | 2014-12-30 | 卡拉ok评分方法和装置 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN105788581B (zh) |
| WO (1) | WO2016095280A1 (zh) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106782600B (zh) * | 2016-12-29 | 2020-04-24 | 广州酷狗计算机科技有限公司 | 音频文件的评分方法及装置 |
| CN107507628B (zh) * | 2017-08-31 | 2021-01-15 | 广州酷狗计算机科技有限公司 | 唱歌评分方法、装置及终端 |
| CN107978322A (zh) * | 2017-11-27 | 2018-05-01 | 北京酷我科技有限公司 | 一种k歌打分算法 |
| CN108008930B (zh) * | 2017-11-30 | 2020-06-30 | 广州酷狗计算机科技有限公司 | 确定k歌分值的方法和装置 |
| CN108010503B (zh) * | 2017-12-27 | 2021-12-03 | 北京酷我科技有限公司 | 一种k歌录音的方法 |
| CN109903779A (zh) * | 2019-01-08 | 2019-06-18 | 北京雷石天地电子技术有限公司 | 一种歌曲演唱评价方法及*** |
| CN109686376B (zh) * | 2019-01-08 | 2020-06-30 | 北京雷石天地电子技术有限公司 | 一种歌曲演唱评价方法及*** |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006031041A (ja) * | 2005-08-29 | 2006-02-02 | Yamaha Corp | フレーズごとに出力される採点データに基づいて採点画像を逐次変化させるカラオケ装置 |
| JP4007298B2 (ja) * | 2003-09-30 | 2007-11-14 | ヤマハ株式会社 | カラオケ装置及びプログラム |
| JP4487209B2 (ja) * | 2006-06-08 | 2010-06-23 | カシオ計算機株式会社 | カラオケ装置およびカラオケ処理プログラム |
| CN101859560A (zh) * | 2009-04-07 | 2010-10-13 | 林文信 | 卡拉ok歌曲伴唱自动评分方法 |
| CN103559875A (zh) * | 2013-10-21 | 2014-02-05 | 福建星网视易信息***有限公司 | 音准抖动修正方法、装置、***及音视频设备和移动终端 |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3293745B2 (ja) * | 1996-08-30 | 2002-06-17 | ヤマハ株式会社 | カラオケ装置 |
| CN101071561A (zh) * | 2006-05-12 | 2007-11-14 | 上海乐金广电电子有限公司 | 卡拉ok音频***的错音矫正方法 |
| JP4926756B2 (ja) * | 2007-02-26 | 2012-05-09 | 株式会社第一興商 | カラオケ効果音出力システム |
| CN101345047B (zh) * | 2007-07-12 | 2012-09-05 | 英业达股份有限公司 | 人声自动校正的混音***及其混音方法 |
| CN102110435A (zh) * | 2009-12-23 | 2011-06-29 | 康佳集团股份有限公司 | 卡拉ok评分方法及*** |
| CN102340482A (zh) * | 2010-07-21 | 2012-02-01 | 崔信奎 | 基于网络的练歌房服务***及利用该服务***的用户终端 |
| GB2498812A (en) * | 2012-01-30 | 2013-07-31 | China Ind Ltd | Providing an time delayed and pitched shifted accompaniment to a sound produced by a user |
| JP2013167828A (ja) * | 2012-02-16 | 2013-08-29 | Sony Corp | 音響処理装置、音響処理方法、プログラム、電子機器、サーバ装置、クライアント装置および音響処理システム |
-
2014
- 2014-12-15 CN CN201410778465.XA patent/CN105788581B/zh active Active
- 2014-12-30 WO PCT/CN2014/095645 patent/WO2016095280A1/zh active Application Filing
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4007298B2 (ja) * | 2003-09-30 | 2007-11-14 | ヤマハ株式会社 | カラオケ装置及びプログラム |
| JP2006031041A (ja) * | 2005-08-29 | 2006-02-02 | Yamaha Corp | フレーズごとに出力される採点データに基づいて採点画像を逐次変化させるカラオケ装置 |
| JP4487209B2 (ja) * | 2006-06-08 | 2010-06-23 | カシオ計算機株式会社 | カラオケ装置およびカラオケ処理プログラム |
| CN101859560A (zh) * | 2009-04-07 | 2010-10-13 | 林文信 | 卡拉ok歌曲伴唱自动评分方法 |
| CN103559875A (zh) * | 2013-10-21 | 2014-02-05 | 福建星网视易信息***有限公司 | 音准抖动修正方法、装置、***及音视频设备和移动终端 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105788581B (zh) | 2019-05-31 |
| CN105788581A (zh) | 2016-07-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2016095280A1 (zh) | 卡拉ok评分方法和装置 | |
| WO2017177524A1 (zh) | 音视频同步播放的方法及装置 | |
| WO2019080406A1 (zh) | 电视机语音交互方法、语音交互控制装置及存储介质 | |
| WO2017005066A1 (zh) | 录制音视频同步时间戳的方法和装置 | |
| WO2020062978A1 (zh) | 一种车内声场调节方法及音频*** | |
| WO2017143692A1 (zh) | 智能电视及其语音控制方法 | |
| WO2017054488A1 (zh) | 电视播放控制方法、服务器及电视播放控制*** | |
| WO2019085543A1 (zh) | 电视机***及电视机控制方法 | |
| WO2019051902A1 (zh) | 终端控制方法、空调器及计算机可读存储介质 | |
| WO2016201745A1 (zh) | 基于用户终端的就诊提示方法、用户终端和网络医院平台 | |
| WO2018045682A1 (zh) | 音画同步测试方法及装置 | |
| WO2017045441A1 (zh) | 基于智能电视的音频播放方法及装置 | |
| WO2019051905A1 (zh) | 空调器控制方法、空调器及计算机可读存储介质 | |
| WO2018032680A1 (zh) | 音视频播放方法及*** | |
| WO2016101698A1 (zh) | 基于dlna技术实现屏幕推送的方法及*** | |
| WO2018023926A1 (zh) | 电视与移动终端的互动方法及*** | |
| WO2018233221A1 (zh) | 多窗口声音输出方法、电视机以及计算机可读存储介质 | |
| WO2019114127A1 (zh) | 空气调节器的语音播报方法及装置 | |
| WO2017088427A1 (zh) | 音频输出控制方法及装置 | |
| WO2017121066A1 (zh) | 应用程序显示方法和*** | |
| WO2016095244A1 (zh) | 视频会议中视频窗口的调整方法及装置 | |
| WO2018090446A1 (zh) | 投影仪调试安装方法及装置 | |
| WO2016090991A1 (zh) | 流媒体数据的下载方法及装置 | |
| WO2014163231A1 (ko) | 복수의 음원이 출력되는 환경하에서 음성 인식에 이용될 음성 신호의 추출 방법 및 음성 신호의 추출 장치 | |
| WO2017080195A1 (zh) | 音频识别方法及装置 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 14908315 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 32PN | Ep: public notification in the ep bulletin as address of the adressee cannot be established |
Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 31.10.2017) |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 14908315 Country of ref document: EP Kind code of ref document: A1 |