CN113255348B - Lyric segmentation method, device, equipment and medium - Google Patents
Lyric segmentation method, device, equipment and medium Download PDFInfo
- Publication number
- CN113255348B CN113255348B CN202110579130.5A CN202110579130A CN113255348B CN 113255348 B CN113255348 B CN 113255348B CN 202110579130 A CN202110579130 A CN 202110579130A CN 113255348 B CN113255348 B CN 113255348B
- Authority
- CN
- China
- Prior art keywords
- lyric
- lyrics
- time
- duration
- chord
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Reverberation, Karaoke And Other Acoustics (AREA)
- Auxiliary Devices For Music (AREA)
Abstract
The application discloses a lyric segmentation method, a device, equipment and a medium, wherein the method comprises the following steps: acquiring lyric data to be processed; determining the time length of each lyric based on the starting time of each lyric of the lyric data to be processed, and determining the single chord time length based on the distribution concentration degree of the time length; the lyrics are segmented by the single chord duration, and the lyrics are divided into different lyric segments. Therefore, after the lyric data to be processed is obtained, the lyrics are directly segmented according to the starting time of each obtained lyric, other training samples except the lyric data to be processed are not needed, and the final lyric segmentation is divided based on the starting time of each lyric, so that the applicability is strong. The lyrics are segmented by using the single chord duration, so that the finally segmented lyrics are segmented based on the music theory knowledge, and the segmented lyrics can be more suitable for the auditory sense of people, thereby improving the user experience and the accuracy rate of the lyrics segmentation.
Description
Technical Field
The present application relates to the field of computer technologies, and in particular, to a method, an apparatus, a device, and a medium for segmenting lyrics.
Background
In the live broadcasting service, two or more than two main broadcasting connecting wheat can carry out song antiphonal singing, one song sings one section by one main broadcasting, and the lyrics loaded from a lyric library need to be reasonably segmented so as to divide the whole lyrics into different lyric fragments, so that the different lyric fragments are distributed to different main broadcasting.
In order to deal with the above situation, in the prior art, a segmentation model is obtained mainly through offline training based on the format of the lyrics, audio characteristics and the like, and then lyric segmentation information is calculated offline by using the segmentation model. In this way, the training of the segmentation model needs to rely on offline calculation, and is limited by the training data, so that all songs cannot be segmented, for example, if the training data is chinese songs, the trained classification model is difficult to correctly segment english songs.
Disclosure of Invention
In view of this, an object of the present application is to provide a method, an apparatus, a device, and a medium for lyric segmentation, which do not depend on training samples, have strong adaptability, and can improve the accuracy rate of lyric segmentation. The specific scheme is as follows:
in a first aspect, the present application discloses a lyric segmentation method, comprising:
acquiring lyric data to be processed, wherein the lyric data to be processed comprises lyrics and the starting time of each lyric;
determining the time length of each sentence of the lyrics based on the starting time of each sentence of the lyrics, and determining the single chord time length corresponding to the lyric data to be processed based on the distribution concentration degree of the time length, wherein the single chord time length is the time length of a chord;
and segmenting the lyrics by utilizing the single chord duration so as to divide the lyrics into different lyric segments.
Optionally, the segmenting the lyrics by using the single chord duration to divide the lyrics into different lyric segments includes:
dividing the lyrics into different lyric large sections based on the single chord duration and the starting time of each sentence of the lyrics;
dividing each of the lyric episodes into different lyric fragments based on the chord length of the chord.
Optionally, the dividing the lyrics into different lyric large sections based on the single chord duration and the starting time of each sentence of the lyrics includes:
determining a time interval between every two adjacent words of lyrics based on a starting time of each word of the lyrics;
judging whether each time interval exceeds a reference interval value, wherein the reference interval value is a value determined based on the single chord duration;
if the time interval exceeds the reference interval value, dividing the former lyric in the two adjacent lyrics corresponding to the time interval from the first lyric to the first lyric exceeding the reference interval value into a large lyric segment, and dividing the former lyric in the two adjacent lyrics corresponding to the time interval from the first lyric to the second lyric to the first lyric to the large lyric segment until the division of the lyrics is finished.
Optionally, dividing any lyric large segment into different lyric small segments based on the chord length of the single chord comprises:
determining a first time range according to the starting time of the first sentence of the lyrics of the large section of the lyrics and the single chord duration;
dividing a first lyric small section of the lyric large section from the lyric large section according to the first time range;
if the division of the lyric large section is not finished, determining a second time range according to the starting time of the next lyric adjacent to the last lyric in the first lyric small section of the lyric large section and the single chord duration;
and dividing a second lyric small section of the lyric large section from the lyric large section according to the second time range until the division of the lyric large section is finished.
Optionally, the determining a first time range according to the starting time of the first sentence of lyrics of the large paragraph of the lyrics and the duration of the chord comprises:
and taking the starting time of the first lyric of the large lyric section as a time lower bound, and taking the sum of the starting time of the first lyric of the large lyric section and the preset number of the single chord time lengths as a time upper bound to obtain the first time range.
Optionally, the dividing the first lyric fragment of the lyric fragment from the lyric fragment according to the first time range includes:
judging whether the ending time of each lyric in the lyric large segment is within the first time range from the first lyric in the lyric large segment, wherein the ending time of each lyric is the sum of the starting time of each lyric and the duration of each lyric;
if the ending time of the lyrics of the current sentence is within the first time range, dividing the lyrics of the current sentence into small lyrics of the first large lyrics of the large lyrics;
and if the ending time of the lyrics of the current sentence is greater than the upper time limit of the first time range, determining the overlapping duration of the lyrics of the current sentence in the first time range, and determining whether to divide the lyrics of the current sentence into a first lyric small section of a large section of the lyrics based on the overlapping duration.
Optionally, the determining an overlap duration of the lyrics of the current sentence in the first time range and determining whether to divide the lyrics of the current sentence into a first lyric small segment of the large segment of the lyrics based on the overlap duration includes:
taking the difference between the time upper bound of the first time range and the starting time of the lyrics of the current sentence as the overlapping duration;
determining the ratio of the overlapping duration to the duration of the lyrics of the current sentence, and judging whether the ratio is not less than a preset ratio threshold value;
if so, the lyrics of the current sentence are divided into the first lyric small section of the large lyric section.
Optionally, the determining the single chord duration corresponding to the lyric data to be processed based on the distribution concentration degree of the durations includes:
sequencing the time lengths of the lyrics of each sentence in a sequence from small to large to obtain a time length sequence;
traversing the time length sequence to divide the time lengths in the time length sequence into different time length groups;
and determining the maximum number of the time lengths in the time length groups as a target time length group with the highest distribution concentration degree, and taking the average value of each time length in the target time length group as the chord time length.
In a second aspect, the present application discloses a lyric segmentation apparatus, comprising:
the system comprises a data acquisition module, a processing module and a processing module, wherein the data acquisition module is used for acquiring lyric data to be processed, and the lyric data to be processed comprises lyrics and the starting time of each sentence of lyrics;
the chord duration determining module is used for determining the duration of each lyric based on the starting time of each lyric and determining the chord duration corresponding to the lyric data to be processed based on the distribution concentration degree of the duration, wherein the chord duration is the duration of a chord;
and the segmentation module is used for segmenting the lyrics by utilizing the chord duration of the single chord so as to divide the lyrics into different lyric segments.
In a third aspect, the present application discloses an electronic device, comprising:
a memory and a processor;
wherein the memory is used for storing a computer program;
the processor is configured to execute the computer program to implement the lyrics segmentation method disclosed above.
In a fourth aspect, the present application discloses a computer readable storage medium for storing a computer program, wherein the computer program when executed by a processor implements the lyrics segmentation method disclosed above.
Therefore, the lyric data to be processed is obtained firstly, wherein the lyric data to be processed comprises lyrics and the starting time of each lyric. And then determining the time length of each sentence of the lyrics based on the starting time of each sentence of the lyrics, and determining the single chord time length corresponding to the lyric data to be processed based on the distribution concentration degree of the time length, wherein the single chord time length is the time length of a chord. And segmenting the lyrics by utilizing the single chord duration so as to divide the lyrics into different lyric segments. Therefore, the lyric data to be processed including the lyrics and the starting time of each lyric are obtained, the time length of each lyric can be determined based on the starting time of each lyric, the single chord time length corresponding to the lyric data to be processed is determined based on the distribution concentration degree of the time length, then the lyrics in the lyric data to be processed are segmented by the determined single chord time length, and the lyrics are divided into different lyric segments. In addition, before segmenting the lyrics, the single chord duration, namely the duration of a chord, needs to be determined, and then the lyrics are segmented by using the single chord duration, so that the finally segmented lyrics are segmented based on the music theory knowledge, the auditory sense of people can be better matched, the user experience can be improved, and the accuracy rate of lyrics segmentation can also be improved.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, it is obvious that the drawings in the following description are only embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
FIG. 1 is a schematic diagram of a system framework to which the lyric segmentation scheme provided herein is applicable;
FIG. 2 is a flow chart of a lyric segmentation method disclosed in the present application;
FIG. 3 is a flow chart of a lyric segmentation method disclosed in the present application;
FIG. 4 is a partial flow diagram of a particular lyric segmentation method disclosed herein;
FIG. 5 is a flow diagram of a portion of a specific lyric segmentation method disclosed herein;
FIG. 6 is a schematic diagram of a lyric segmentation apparatus disclosed in the present application;
fig. 7 is a schematic structural diagram of an electronic device disclosed in the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
For ease of understanding, a system framework to which the lyric segmentation method of the present application is applicable will be described. It should be understood that, in the embodiment of the present application, the number of the computer devices is not limited, and it may be that multiple computer devices cooperate together to perform the lyric segmentation function. In one possible scenario, please refer to fig. 1. As can be seen from fig. 1, the hardware composition framework may include: a first computer device 101, a second computer device 102. The first computer device 101 and the second computer device 102 are communicatively connected via a network 103.
In the embodiment of the present application, the hardware structures of the first computer device 101 and the second computer device 102 are not specifically limited, and the first computer device 101 and the second computer device 102 perform data interaction to implement the lyric segmentation function. Further, the form of the network 103 is not limited in this embodiment, for example, the network 103 may be a wireless network (e.g., WIFI, bluetooth, etc.), or may be a wired network.
The first computer device 101 and the second computer device 102 may be the same computer device, for example, the first computer device 101 and the second computer device 102 are both servers; or different types of computer devices, e.g., the first computer device 101 may be a terminal or an intelligent electronic device, and the second computer device 102 may be a server. In yet another possible scenario, a server with high computing power may be utilized as the second computer device 102 to improve data processing efficiency and reliability, and thus lyric segmentation efficiency. Meanwhile, a terminal or an intelligent electronic device with low cost and wide application range is used as the first computer device 101 to realize the interaction between the second computer device 102 and the user.
For example, referring to fig. 2, when a terminal acquires a song antiphonal singing request triggered in a live broadcast service, a terminal sends related information of a song to be antiphonal singing to a corresponding server, and the server loads lyric data to be processed of the song to be antiphonal singing according to the received related information of the song to be antiphonal singing, wherein the lyric data to be processed includes lyrics and start time of each sentence of lyrics. And then the server also needs to determine the single chord time length corresponding to the lyric data to be processed based on the time length of each lyric at the starting time of each lyric and the distribution concentration degree of the time length, wherein the single chord time length is the time length of one chord, then the lyrics in the lyric data to be processed are segmented by utilizing the single chord time length, the lyrics are divided into different lyric segments, and then the lyric segmentation result is returned to the terminal, so that the terminal can display the segmented lyrics in a lyric display area after receiving the lyric segmentation result.
Referring to fig. 3, an embodiment of the present application discloses a lyric segmentation method, including:
step S11: obtaining lyric data to be processed, wherein the lyric data to be processed comprises lyrics and the starting time of each lyric.
In a specific implementation process, lyric data to be processed needs to be acquired first, wherein the lyric data to be processed includes lyrics and the starting time of each sentence of lyrics, and a sentence of lyrics is also a line of lyrics in the lyric data. The lyrics to be processed may be LRC (lyrics), which is also called as LRC lyrics data, and the LRC lyrics are currently general player lyrics data. For example, LRC lyrics are as follows, where one word is counted per line of lyrics, and the number in parentheses before each word is the start time of the word.
[ 00.00] simple happiness
[00
[00
[00
[00
[00
[00
The lyric data to be processed may also be QRC lyrics, the QRC lyrics may be player lyric data dedicated to the QQ music, the QRC lyrics may include lyrics and a start time of each lyric, and may further include a reference time length of each lyric, for example, the QRC lyrics are as follows, where each line of lyrics is counted as one sentence, a number before a bracket comma in front of each lyric is the start time of the lyric, the start time is in units of milliseconds, and a number after the comma is the reference time length of the lyric.
[4075,3897] the yellowish spring festival scrolls remain on the wall
[8242,3737] is rare and can be seen as peace of life
[12246,3913] old household rice jar that I did not go back
[16479,3784] grandpa uses regular script to write one full
Step S12: and determining the time length of each lyric based on the starting time of each lyric, and determining the single chord time length corresponding to the lyric data to be processed based on the distribution concentration degree of the time length, wherein the single chord time length is the time length of a chord.
After the lyric data to be processed is obtained, the time length of each sentence of the lyrics is determined based on the starting time of each sentence of the lyrics, and the single chord time length corresponding to the lyric data to be processed is determined based on the distribution concentration degree of the time length, wherein the single chord time length is the time length of a chord. Since a song segment is a set of chords from the music point of view, the song needs to be reasonably segmented and needs to be matched with the sense of a person. Therefore, it is necessary to determine the duration of a set of chords, and a set of chords usually consists of 4 chords, so that the duration of a chord needs to be determined first, and usually, one chord is used for one or two musical bars, that is, the duration of one chord is determined, and the duration of the highest concentration of the time durations of each lyric is generally the duration of one or two musical bars.
That is, it is necessary to determine the time length of each lyric based on the starting time of each lyric, where the time length of each lyric is different from the reference time length of each lyric included in the lyric data when the lyric data to be processed is QRC lyrics, where the time length of each lyric is generally greater than the reference time length of each lyric included in the lyric data when the lyric data to be processed is QRC lyrics, because a singer may need to pause for a short period of time after a lyric sings to exchange air, etc. After the time length of each lyric is determined based on the starting time of each lyric, the single chord time length corresponding to the lyric data to be processed is determined based on the distribution concentration degree of the time length.
Wherein determining the duration of each sentence of the lyrics based on the starting time of each sentence of the lyrics comprises: and taking the value of the difference between the starting time of any one lyric and the starting time of the next lyric adjacent to the lyric as the time length of the lyric until the time length of each lyric is determined.
For example, the first sentence of lyrics is: [4075,3897] the yellowish spring festival scrolls also remained on the wall, with the second words of lyrics: [8242,3737] it is rare to see that the words are peaceful all the year round, and the duration of the first word "the yellow spring festival still remains on the wall" is 8242-4075=4167 ms.
Determining the single chord duration corresponding to the lyric data to be processed based on the distribution concentration degree of the duration, wherein the method comprises the following steps:
step S121: and sequencing the time lengths of the lyrics of each sentence in a sequence from small to large to obtain a time length sequence.
When the single chord time length corresponding to the lyric data to be processed is determined based on the distribution concentration degree of the time length, the time lengths of each sentence of the lyrics need to be sequenced from small to large to obtain a time length sequence, and the sequence of the time lengths of each sentence of the lyrics from small to large is favorable for traversing processing of the next step.
Step S122: and traversing the time length sequence to divide the time lengths in the time length sequence into different time length groups.
After the duration sequence is obtained, traversing the duration sequence is needed to divide the durations in the duration sequence into different duration groups. The time length sequence can be traversed for multiple times by taking different time length thresholds as centers and corresponding time length deviations as deviation values.
For example, the duration of each sentence of the lyrics is: 1. 200, 29, 3, 201, 215, and 208, the time lengths of the lyrics are arranged in the order from small to large, and the obtained time length sequence is as follows: 1.3, 29, 200, 201, 208, 215, wherein 50 is taken as the center in the first traversal, plus or minus 50 is the deviation to bind the values in the duration sequence, that is, the duration within the range of binding 0-100, to obtain a first duration subgroup of 1,3, 29, 100 is taken as the center in the second traversal, plus or minus 50 is the deviation to bind the values in the duration sequence, that is, the duration within the range of binding 50-150, to obtain a second duration subgroup of null, 150 is taken as the center in the third traversal, plus or minus 50 is the deviation to bind the values in the duration sequence, that is, the duration within the range of binding 100-200, to obtain a third duration subgroup of 200, 200 is taken as the center in the fourth traversal 200, 50 is the deviation to bind the values in the duration sequence, that is, the duration within the range of binding 200-250, to obtain a fourth duration subgroup of 200, 201, 208, 215. The same time length may belong to different time length groups.
In an actual implementation process, the deviations may be different according to different values of the center, for example, in the above process, 50 may be used as the center for the first traversal, plus or minus 50 is used as the deviation to bind the values in the duration sequence, that is, bind the duration within a range of 0 to 100, to obtain first duration groups of 1,3, and 29, the second traversal 100 is used as the center, plus or minus 70 is used as the deviation to bind the values in the duration sequence, that is, bind the duration within a range of 30 to 170, to obtain a second duration group of null, the third traversal 150 is used as the center, plus or minus 90 is used as the deviation to bind the values in the duration sequence, that is, bind the duration within a range of 60 to 240, to obtain a third duration group of 200, 201, 208, and 215, and since there is no duration greater than 240 in the duration sequence, the fourth traversal is not needed.
Step S123: and determining the maximum number of the time lengths in the time length groups as a target time length group with the highest distribution concentration degree, and taking the average value of each time length in the target time length group as the chord time length.
And after obtaining each time length group, determining that the time length group with the largest number of the time lengths is the target time length group with the highest distribution concentration degree, and taking the average value of each time length in the target time length group as the chord time length. For example, if the group with the largest number of time lengths in the time length groups is 200, 201, 208, and 215, the average value thereof is 206 as the corresponding chord time length.
Step S13: and segmenting the lyrics by utilizing the chord duration of the single chord so as to divide the lyrics into different lyric segments.
After the chord duration is determined, the chord duration in the lyric data to be processed is also determined, and then the lyric can be segmented by using the chord duration to divide the lyric into different lyric segments.
Specifically, the division of the whole lyric may be performed in two steps, that is, the lyric may be divided into different lyric large sections based on the single chord duration and the starting time of each sentence of the lyric; dividing each of the lyric large sections into different lyric small sections based on the chord length of the single chord. Because the time interval between the large lyric fragments is long, generally speaking, the lyrics in different large lyric fragments need to be divided into different small lyric fragments, so the lyrics of the whole song are divided into different large lyric fragments, and then each large lyric fragment is divided into different small lyric fragments.
Therefore, the lyric data to be processed is obtained firstly, wherein the lyric data to be processed comprises lyrics and the starting time of each lyric. And then determining the time length of each lyric based on the starting time of each lyric, and determining the single chord time length corresponding to the lyric data to be processed based on the distribution concentration degree of the time length, wherein the single chord time length is the time length of a chord. And segmenting the lyrics by utilizing the single chord duration so as to divide the lyrics into different lyric segments. Therefore, the lyric data to be processed including the lyrics and the starting time of each lyric is obtained, the time length of each lyric can be determined based on the starting time of each lyric, the single chord time length corresponding to the lyric data to be processed is determined based on the distribution concentration degree of the time length, the lyrics in the lyric data to be processed are segmented by the determined single chord time length, and the lyrics are divided into different lyric segments. In addition, before segmenting the lyrics, the single chord duration, namely the duration of a chord, needs to be determined, and then the lyrics are segmented by using the single chord duration, so that the finally segmented lyrics are segmented based on the music theory knowledge, the auditory sense of people can be better matched, the user experience can be improved, and the accuracy rate of lyrics segmentation can also be improved.
Referring to fig. 4, the dividing the lyrics into different lyric large sections based on the chord length of the chord and the start time of each sentence of the lyrics includes:
step S21: the time interval between every two adjacent words of lyrics is determined based on the starting time of each said word of lyrics.
The whole lyrics need to be divided into different lyric large sections according to the time interval between each sentence of lyrics, so the time interval between every two adjacent sentences of lyrics needs to be determined based on the starting time of each sentence of the lyrics. Since the LRC lyrics need to be considered to have only the starting time of each lyric, the time interval between every two adjacent lyrics is determined based on the starting time of each lyric, and the value of the difference between the starting time of any one lyric and the starting time of the next lyric adjacent to the lyric can be used as the time interval between the lyric and the next lyric adjacent to the lyric until the time interval between every two adjacent lyrics is determined.
Step S22: and judging whether each time interval exceeds a reference interval value, wherein the reference interval value is a value determined based on the single chord duration.
After the time interval between every two adjacent lyrics is determined, it is further required to determine whether each time interval exceeds a reference interval value, where the reference interval value is a value determined based on the single chord duration, specifically, the reference interval value may be two single chord durations, for example, if the single chord duration is 3000 milliseconds, the reference interval value may be 6000 milliseconds. The reference interval value may be determined according to actual conditions, and is not specifically limited herein.
Step S23: if the time interval exceeds the reference interval value, dividing the previous lyric in two adjacent lyrics corresponding to the time interval from the first lyric to the first lyric exceeding the reference interval value into a large lyric segment, and dividing the next lyric in two adjacent lyrics corresponding to the time interval exceeding the reference interval value to the second lyric in two adjacent lyrics exceeding the reference interval value into a large lyric segment until the lyrics are divided.
If the time interval exceeds the reference interval value, dividing the former lyric in the two adjacent lyrics corresponding to the time interval from the first lyric to the first lyric exceeding the reference interval value into a large lyric segment, dividing the latter lyric in the two adjacent lyrics corresponding to the time interval exceeding the reference interval value to the former lyric in the two adjacent lyrics corresponding to the second lyric exceeding the reference interval value into a large lyric segment, and so on until the division of the lyrics is finished.
Referring to fig. 5, dividing any one lyric large section into different lyric small sections based on the chord length of the chord comprises:
step S31: and determining a first time range according to the starting time of the first sentence of the lyric of the large paragraph of the lyric and the single chord duration.
In the process of dividing any lyric large segment into different lyric small segments based on the single chord duration, a first time range is determined according to the starting time of the first sentence of lyrics of the lyric large segment and the single chord duration.
Specifically, the first time range is obtained by taking the starting time of the first lyric of the large lyric segment as a time lower bound and taking the sum of the starting time of the first lyric of the large lyric segment and the preset number of single chord durations as a time upper bound. Wherein the predetermined number is typically 4, since the set of chords is typically 4 chords. For example, the starting time of the lyric of the first sentence of the large lyric segment is 3000 milliseconds, the single chord duration is 3500 milliseconds, and the first time range is 3000-17000 milliseconds.
Step S32: and dividing a first lyric small section of the lyric large section from the lyric large section according to the first time range.
After the first time range is determined, a first lyric fragment of the lyric fragment can be divided from the lyric fragment according to the first time range.
Specifically, whether the ending time of each lyric in the large lyric segment is within the first time range is judged from the first lyric in the large lyric segment, wherein the ending time of each lyric is the sum of the starting time of each lyric and the duration of each lyric; if the ending time of the lyrics of the current sentence is within the first time range, dividing the lyrics of the current sentence into a first lyrics small section of a large section of the lyrics; and if the ending time of the lyrics of the current sentence is greater than the upper time limit of the first time range, determining the overlapping duration of the lyrics of the current sentence in the first time range, and determining whether to divide the lyrics of the current sentence into a first lyric small section of a large section of the lyrics based on the overlapping duration.
Firstly, whether the ending time of each lyric in the lyric large segment is in the first time range is judged from the first lyric in the lyric large segment, the ending time of each lyric is the sum of the starting time of each lyric and the time length of each lyric, when the lyric data to be processed is LRC lyrics, the difference value between the starting time of any lyric and the starting time of the next lyric adjacent to the lyric can be used as the time length of the lyric until the time length of each lyric is determined, and the time length of each lyric is stored so that the sum of the starting time of each lyric and the time length of each lyric is used as the corresponding ending time. When the lyric data to be processed is QRC lyrics, the sum of the starting time of each lyric and the reference time length of each lyric may be used as the corresponding ending time. If the ending time of the lyrics of the current sentence is within the first time range, the whole time of the lyrics of the current sentence is within the first time range, so that the lyrics of the current sentence are divided into a small lyric segment of the large lyric segment. If the ending time of the lyric of the current sentence is larger than the upper time limit of the first time range, the fact that the time of the lyric of the current sentence may be partially in the first time range is indicated, further determination is needed to be made whether the lyric of the current sentence is divided into a first lyric small section of the large lyric segment, so that the overlapping time length of the lyric of the current sentence in the first time range needs to be determined, and whether the lyric of the current sentence is divided into the first lyric small section of the large lyric segment is determined based on the overlapping time length.
Determining the overlapping time length of the lyrics of the current sentence in the first time range, and determining whether to divide the lyrics of the current sentence into a first lyric small section of a large lyric section based on the overlapping time length, wherein the determining comprises the following steps: taking the difference between the time upper bound of the first time range and the starting time of the lyrics of the current sentence as the overlapping duration; determining the ratio of the overlapping duration to the duration of the lyrics of the current sentence, and judging whether the ratio is not less than a preset ratio threshold value; if so, the lyrics of the current sentence are divided into the first lyric small section of the large lyric section. Wherein, the preset ratio threshold may be 60%.
For example, the first time range is 3000-17000 ms, the starting time of the lyric of the current sentence is 14000 ms, the ending time of the lyric of the current sentence is 18000 ms, and the preset ratio threshold may be 60%. The overlapping duration is 17000-14000=3000 milliseconds, the duration of the lyrics of the current sentence is 4000 milliseconds, and the ratio of the overlapping duration to the duration of the lyrics of the current sentence is: 3000/4000=0.75, since 0.75 is larger than 0.6, so divide the lyrics of the current sentence into the first lyrics small fragment of the large fragment of the lyrics.
Step S33: if the division of the lyric large section is not finished, a second time range is determined according to the starting time of the next lyric adjacent to the last lyric in the first lyric small section of the lyric large section and the single chord time length.
The start time of the second lyric fragment in the lyric large fragment needs to be determined after the division of the first lyric fragment in the lyric large fragment is finished, and if the division of the lyric large fragment is not finished, a second time range is determined according to the start time of the next lyric adjacent to the last lyric in the first lyric fragment of the lyric large fragment and the single chord duration.
Specifically, the second time range is obtained by taking the starting time of the next lyric adjacent to the last lyric in the first lyric small section of the large lyric segment as a lower time boundary, and taking the sum of the starting time of the next lyric adjacent to the last lyric in the first lyric small section of the large lyric segment and the preset number of single chord time durations as an upper time boundary.
Step S34: and dividing a second lyric small section of the lyric large section from the lyric large section according to the second time range until the division of the lyric large section is finished.
After the second time range is obtained, a second lyric small section of the lyric large section can be divided from the lyric large section according to the second time range until the division of the lyric large section is finished.
The foregoing description in the present application is illustrated by taking QRC formatted lyrics of shanghai-nine-four-three as an example, and the specific lyrics data are as follows:
[4075,3897] the yellowish spring festival scrolls also remained on the walls; [8242,3737] is rare and has a few characters and peace all the year round; [12246,3913] old rice jar I did not go back; [16479,3784] grandpa was written in regular script with one full; [20606,3864] gold kudzuvine climbs over the carved door and window; [24734,3807] the oblique sun is reflected on the mottled brick wall; [28807,3831] beech boards are spread in the house; the soybean paste is brewed in the current year in the [32973,4025] grandma; [37351,3656] I begin imagining against a black and white photograph; [41266,4514] the current year's patterns for dad and mom; [46687,3695] a girl speaking a bite of the Rongnong Soft language; [50382,3896] walk slowly over the beach; [54831,2241] evanescent old time light; [57072,1463] one, nine, four, three; [58815,2400] on the way to recall; [61215,1712] time becomes slow and good; [63318,2073] old archway small arcade; [65391,3752] is white wall black tile in that age; [69143,1880] minor concerns; [71246,2393] evanescent old time light; [73639,1327] one, nine, four, three; [75391,2392] fragment looking back at; [77783,1745] there is some wind frost; [79820,2121] old CD case; [81941,3415] in a tin filled with postcards; [85356,3896] harbor a rose petal; [104491,3784] gold kudzuvine climbs over the carved door and window; [108515,3856] the oblique sun is reflected on the mottled brick wall; [112693,3800] beech boards are spread in the house; the soybean paste is brewed in the current year in the [116831,4064] grandma; [121191,3617] I begin imagining against a black and white photograph; [125064,4592] the current year's patterns for dad and mom; [130551,3463] a girl speaking Wunong Soft language Yi kou; [134014,5539] walk slowly over the beach; [139553,2361] evanescent old time light; [141914,1455] one, nine, four, three; [143649,2416] on the recalled way; [146065,1768] time becomes slow and good; [148078,2136] old archway small arcade; [150214,3776] is of white wall black tile in that age; [153990,1880] minor concerns; [156204,2256] evanescent old time light; [158460,1448] one, nine, four, three; [160228,2376] fragment looking back at; [162604,1656] there is some wind frost; [164575,2224] old CD case; [166799,3540] filled in a tin of postcard; [170339,8016] harbor one rose petal.
Wherein, one semicolon in the lyrics of Shanghai-nine-four-three represents the original one line of lyrics, and one line of lyrics is also one sentence of lyrics, and the value of the difference between the starting time of any one sentence of lyrics and the starting time of the next sentence of lyrics adjacent to the one sentence of lyrics is used as the duration of the one sentence of lyrics until the duration of each sentence of lyrics is determined. Then, the time lengths of the lyrics of each sentence are sequenced from small to large to obtain a time length sequence, the time length sequence is traversed to divide the time lengths in the time length sequence into different time length groups, and specifically, the small time length group information is as follows: 3873, 3915, 4004, 4024, 4073, 4127, 4128, 4138, 4166, 4167, 4178, 4233; [ 2103, 2121, 2136, 2214, 2224, 2241, 2256 ]; [ 3415, 3463, 3540, 3695, 3752, 3776 ]; [ 2361, 2376, 2392, 2393, 2400, 2416 ]; [ 1735, 1743, 1752, 1768 ]. One middle bracket represents one time length subgroup, so that the time length subgroup with the largest number of time lengths is represented as [ 3873, 3915, 4004, 4024, 4073, 4127, 4128, 4138, 4166, 4167, 4178 and 4233 ], and the average value of each time length value in the time length subgroup is used as the single chord time length of the lyrics of 'one, nine and three in Shanghai' as 4085 milliseconds. Then, twice the chord duration is used as a reference interval value, namely the reference interval value is 8170 milliseconds, the lyrics of Shanghai-nine-four-three are divided into large sections, the lyrics of the first sentence to the door and window with [104491,3784] golden kudzuvine climbing over the sculpture "can be divided into a lyrics large section, and the lyrics of the first sentence to the door and window with [108515,3856] sunset obliquely reflected on the mottled brick wall" can be divided into another large section. Dividing the first lyric large segment into small lyric segments, wherein the preset number is 4, and the first time range of the first lyric large segment is 4075-20415 milliseconds, so that the "[4075,3897] yellowed spring festival scrolls can be remained on the wall; [8242,3737] is rare and has been seen to be peace of life for several years; [12246,3913] old rice jar I did not go back; [16479,3784] the grandpa is divided into the first lyric fragments in the first lyric fragment by writing a full in regular script, and so on, the whole song can be divided into different lyric fragments.
Referring to fig. 6, an embodiment of the present application discloses a lyric segmentation apparatus, including:
the data acquisition module 21 is configured to acquire lyric data to be processed, where the lyric data to be processed includes lyrics and a start time of each sentence of lyrics;
a chord duration determining module 22, configured to determine a duration of each lyric based on a start time of each lyric, and determine a chord duration corresponding to the lyric data to be processed based on a distribution concentration degree of the durations, where the chord duration is a chord duration;
a segmenting module 23, configured to segment the lyrics by using the chord duration to divide the lyrics into different lyric segments.
Therefore, the lyric data to be processed is obtained firstly, wherein the lyric data to be processed comprises lyrics and the starting time of each lyric. And then determining the time length of each lyric based on the starting time of each lyric, and determining the single chord time length corresponding to the lyric data to be processed based on the distribution concentration degree of the time length, wherein the single chord time length is the time length of a chord. And segmenting the lyrics by utilizing the single chord duration so as to divide the lyrics into different lyric segments. Therefore, the lyric data to be processed including the lyrics and the starting time of each lyric are obtained, the time length of each lyric can be determined based on the starting time of each lyric, the single chord time length corresponding to the lyric data to be processed is determined based on the distribution concentration degree of the time length, then the lyrics in the lyric data to be processed are segmented by the determined single chord time length, and the lyrics are divided into different lyric segments. In addition, before segmenting the lyrics, the single chord duration, namely the duration of a chord, needs to be determined, and then the lyrics are segmented by using the single chord duration, so that the finally segmented lyrics are segmented based on the music theory knowledge, the auditory sense of people can be better matched, the user experience can be improved, and the accuracy rate of lyrics segmentation can also be improved.
In some specific implementations, the segmentation module 23 is configured to:
dividing the lyrics into different lyric large sections based on the single chord time length and the starting time of each sentence of the lyrics;
dividing each of the lyric large sections into different lyric small sections based on the chord length of the single chord.
In some specific implementations, the segmentation module 23 is configured to:
determining a time interval between every two adjacent words of lyrics based on the starting time of each word of the lyrics;
judging whether each time interval exceeds a reference interval value, wherein the reference interval value is a value determined based on the single chord duration;
if the time interval exceeds the reference interval value, dividing the previous lyric in two adjacent lyrics corresponding to the time interval from the first lyric to the first lyric exceeding the reference interval value into a large lyric segment, and dividing the next lyric in two adjacent lyrics corresponding to the time interval exceeding the reference interval value to the second lyric in two adjacent lyrics exceeding the reference interval value into a large lyric segment until the lyrics are divided.
In some specific implementations, the segmentation module 23 is configured to:
determining a first time range according to the starting time of the first sentence of the lyrics of the large section of the lyrics and the single chord duration;
dividing a first lyric small section of the lyric large section from the lyric large section according to the first time range;
if the division of the lyric large section is not finished, determining a second time range according to the starting time of the next lyric adjacent to the last lyric in the first lyric small section of the lyric large section and the single chord time length;
and dividing a second lyric small section of the lyric large section from the lyric large section according to the second time range until the division of the lyric large section is finished.
In some specific implementations, the segmentation module 23 is configured to:
and taking the starting time of the first lyric of the large lyric segment as a time lower bound, and taking the sum of the starting time of the first lyric of the large lyric segment and the preset number of the single chord durations as a time upper bound to obtain the first time range.
In some specific implementations, the segmentation module 23 is configured to:
judging whether the ending time of each lyric in the lyric large segment is within the first time range from the first lyric in the lyric large segment, wherein the ending time of each lyric is the sum of the starting time of each lyric and the duration of each lyric;
if the ending time of the lyrics of the current sentence is within the first time range, dividing the lyrics of the current sentence into small lyrics of the first large lyrics of the large lyrics;
and if the ending time of the lyrics of the current sentence is greater than the upper time limit of the first time range, determining the overlapping duration of the lyrics of the current sentence in the first time range, and determining whether to divide the lyrics of the current sentence into a first lyric small section of a large section of the lyrics based on the overlapping duration.
In some specific implementations, the segmentation module 23 is configured to:
taking the difference between the time upper bound of the first time range and the starting time of the lyrics of the current sentence as the overlapping duration;
determining the ratio of the overlapping duration to the duration of the lyrics of the current sentence, and judging whether the ratio is not less than a preset ratio threshold value;
if so, the lyrics of the current sentence are divided into the first lyric small section of the large lyric section.
In some specific implementations, the single chord duration determining module 22 is configured to:
sequencing the time lengths of the lyrics of each sentence in a sequence from small to large to obtain a time length sequence;
traversing the time length sequence to divide the time lengths in the time length sequence into different time length groups;
and determining the maximum number of the time lengths in the time length groups as a target time length group with the highest distribution concentration degree, and taking the average value of each time length in the target time length group as the chord time length.
Fig. 7 is a schematic structural diagram of an electronic device 30 according to an embodiment of the present disclosure, where the user terminal may specifically include, but is not limited to, a smart phone, a tablet computer, a notebook computer, or a desktop computer.
In general, the electronic device 30 in the present embodiment includes: a processor 31 and a memory 32.
The processor 31 may include one or more processing cores, such as a four-core processor, an eight-core processor, and so on. The processor 31 may be implemented by at least one hardware of a DSP (digital signal processing), an FPGA (field-programmable gate array), and a PLA (programmable logic array). The processor 31 may also include a main processor and a coprocessor, where the main processor is a processor for processing data in an awake state, and is also called a Central Processing Unit (CPU); a coprocessor is a low power processor for processing data in a standby state. In some embodiments, the processor 31 may be integrated with a GPU (graphics processing unit) which is responsible for rendering and drawing images to be displayed on the display screen. In some embodiments, the processor 31 may include an AI (artificial intelligence) processor for processing computing operations related to machine learning.
Memory 32 may include one or more computer-readable storage media, which may be non-transitory. Memory 32 may also include high speed random access memory, as well as non-volatile memory, such as one or more magnetic disk storage devices, flash memory storage devices. In this embodiment, the memory 32 is at least used for storing a computer program 321, wherein after being loaded and executed by the processor 31, the computer program is capable of implementing the steps of the lyric segmentation method disclosed in any of the previous embodiments.
In some embodiments, the electronic device 30 may further include a display 33, an input/output interface 34, a communication interface 35, a sensor 36, a power source 37, and a communication bus 38.
Those skilled in the art will appreciate that the configuration shown in FIG. 7 is not limiting to electronic device 30 and may include more or fewer components than those shown.
Further, the present application also discloses a computer-readable storage medium for storing a computer program, wherein the computer program is used for implementing the lyric segmentation method disclosed in any one of the foregoing embodiments when being executed by a processor.
For the specific process of the lyric segmentation method, reference may be made to the corresponding contents disclosed in the foregoing embodiments, and details are not repeated here.
The embodiments are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same or similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module may reside in Random Access Memory (RAM), memory, read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
Finally, it is further noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a series of processes, methods, articles, or apparatus that comprises other elements does not include only those elements but may include other elements not expressly listed or inherent to such processes, methods, articles, or apparatus. Without further limitation, an element defined by the phrase "comprising a … …" does not exclude the presence of another identical element in a process, method, article, or apparatus that comprises the element.
The method, apparatus, device and medium for lyric segmentation provided by the present application are described in detail above, and specific examples are applied herein to explain the principle and the implementation of the present application, and the description of the above embodiments is only used to help understand the method and the core idea of the present application; meanwhile, for a person skilled in the art, according to the idea of the present application, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present application.
Claims (9)
1. A method of lyric segmentation, comprising:
acquiring lyric data to be processed, wherein the lyric data to be processed comprises lyrics and the starting time of each lyric;
determining the time length of each sentence of the lyrics based on the starting time of each sentence of the lyrics, and determining the single chord time length corresponding to the lyric data to be processed based on the distribution concentration degree of the time length, wherein the single chord time length is the time length of a chord;
segmenting the lyrics by utilizing the single chord duration to divide the lyrics into different lyric segments;
wherein the segmenting the lyrics by the chord duration of the single chord to divide the lyrics into different lyric segments comprises:
dividing the lyrics into different lyric large sections based on the single chord duration and the starting time of each sentence of the lyrics;
dividing each lyric large section into different lyric small sections based on the single chord time length;
wherein the dividing of the lyrics into different lyric large sections based on the chord length of the single chord and the starting time of each sentence of the lyrics comprises:
determining a time interval between every two adjacent words of lyrics based on a starting time of each word of the lyrics;
the lyrics are divided into different lyric large sections based on the chord duration and the respective time intervals.
2. The lyric segmentation method of claim 1 wherein said dividing the lyrics into different lyric large sections based on the solo chord duration and the respective time intervals comprises:
judging whether each time interval exceeds a reference interval value, wherein the reference interval value is a value determined based on the single chord duration;
if the time interval exceeds the reference interval value, dividing the previous lyric in two adjacent lyrics corresponding to the time interval from the first lyric to the first lyric exceeding the reference interval value into a large lyric segment, and dividing the next lyric in two adjacent lyrics corresponding to the time interval exceeding the reference interval value to the second lyric in two adjacent lyrics exceeding the reference interval value into a large lyric segment until the lyrics are divided.
3. The lyric segmentation method of claim 1, wherein dividing any lyric large segment into different lyric small segments based on the single chord duration comprises:
determining a first time range according to the starting time of the first sentence of the lyrics of the large section of the lyrics and the single chord duration;
dividing a first lyric small section of the lyric large section from the lyric large section according to the first time range;
if the division of the lyric large section is not finished, determining a second time range according to the starting time of the next lyric adjacent to the last lyric in the first lyric small section of the lyric large section and the single chord time length;
and dividing a second lyric small section of the lyric large section from the lyric large section according to the second time range until the division of the lyric large section is finished.
4. The lyric segmentation method of claim 3 wherein said determining a first time range from a start time of a first sentence of lyrics of the large paragraph of lyrics and said solo chord duration comprises:
and taking the starting time of the first lyric of the large lyric segment as a time lower bound, and taking the sum of the starting time of the first lyric of the large lyric segment and the preset number of the single chord durations as a time upper bound to obtain the first time range.
5. The lyric segmentation method of claim 4 wherein said dividing the first lyric fragment of the lyric fragment from the lyric fragment according to said first time range comprises:
judging whether the ending time of each lyric in the lyric large segment is within the first time range from the first lyric in the lyric large segment, wherein the ending time of each lyric is the sum of the starting time of each lyric and the duration of each lyric;
if the ending time of the lyrics of the current sentence is within the first time range, dividing the lyrics of the current sentence into a first lyrics small section of a large section of the lyrics;
and if the ending time of the lyrics of the current sentence is greater than the upper time limit of the first time range, determining the overlapping duration of the lyrics of the current sentence in the first time range, and determining whether to divide the lyrics of the current sentence into small lyrics of a first large lyric section based on the overlapping duration.
6. The method of claim 5, wherein determining an overlap duration of the lyrics of the current sentence in the first time range and determining whether to divide the lyrics of the current sentence into a first lyric fragment of the large fragment of lyrics based on the overlap duration comprises:
taking the difference value between the time upper bound of the first time range and the starting time of the lyrics of the current sentence as the overlapping duration;
determining the ratio of the overlapping duration to the duration of the lyrics of the current sentence, and judging whether the ratio is not less than a preset ratio threshold value;
if so, the lyrics of the current sentence are divided into the first lyric small section of the large lyric section.
7. The lyric segmentation method according to any one of claims 1 to 6, wherein the determining the chord duration corresponding to the lyric data to be processed based on the degree of the distribution concentration of the duration comprises:
sequencing the time lengths of the lyrics of each sentence in a sequence from small to large to obtain a time length sequence;
traversing the time length sequence to divide the time lengths in the time length sequence into different time length groups;
and determining the maximum number of the time lengths in the time length groups as a target time length group with the highest distribution concentration degree, and taking the average value of each time length in the target time length group as the chord time length.
8. An electronic device, comprising:
a memory and a processor;
wherein the memory is used for storing a computer program;
the processor for executing the computer program to implement the lyrics segmentation method of any of claims 1 to 7.
9. A computer-readable storage medium for storing a computer program, wherein the computer program when executed by a processor implements the lyrics segmentation method of any one of claims 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110579130.5A CN113255348B (en) | 2021-05-26 | 2021-05-26 | Lyric segmentation method, device, equipment and medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110579130.5A CN113255348B (en) | 2021-05-26 | 2021-05-26 | Lyric segmentation method, device, equipment and medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113255348A CN113255348A (en) | 2021-08-13 |
CN113255348B true CN113255348B (en) | 2023-02-28 |
Family
ID=77184585
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110579130.5A Active CN113255348B (en) | 2021-05-26 | 2021-05-26 | Lyric segmentation method, device, equipment and medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113255348B (en) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102467939A (en) * | 2010-11-04 | 2012-05-23 | 北京彩云在线技术开发有限公司 | Song audio frequency cutting apparatus and method thereof |
CN103714806A (en) * | 2014-01-07 | 2014-04-09 | 天津大学 | Chord recognition method combining SVM with enhanced PCP |
CN107452361A (en) * | 2017-08-08 | 2017-12-08 | 腾讯音乐娱乐(深圳)有限公司 | Song subordinate sentence method and device |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2011215358A (en) * | 2010-03-31 | 2011-10-27 | Sony Corp | Information processing device, information processing method, and program |
CN105187936B (en) * | 2015-06-15 | 2018-08-21 | 福建星网视易信息***有限公司 | Based on the method for broadcasting multimedia file and device for singing audio scoring |
US20180366097A1 (en) * | 2017-06-14 | 2018-12-20 | Kent E. Lovelace | Method and system for automatically generating lyrics of a song |
-
2021
- 2021-05-26 CN CN202110579130.5A patent/CN113255348B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102467939A (en) * | 2010-11-04 | 2012-05-23 | 北京彩云在线技术开发有限公司 | Song audio frequency cutting apparatus and method thereof |
CN103714806A (en) * | 2014-01-07 | 2014-04-09 | 天津大学 | Chord recognition method combining SVM with enhanced PCP |
CN107452361A (en) * | 2017-08-08 | 2017-12-08 | 腾讯音乐娱乐(深圳)有限公司 | Song subordinate sentence method and device |
Also Published As
Publication number | Publication date |
---|---|
CN113255348A (en) | 2021-08-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107123415B (en) | Automatic song editing method and system | |
CN104143327B (en) | A kind of acoustic training model method and apparatus | |
US20200005761A1 (en) | Voice synthesis method, apparatus, device and storage medium | |
KR102128926B1 (en) | Method and device for processing audio information | |
CN101452696B (en) | Signal processing device, signal processing method and program | |
US11511200B2 (en) | Game playing method and system based on a multimedia file | |
CN106157979B (en) | A kind of method and apparatus obtaining voice pitch data | |
CN110503944B (en) | Method and device for training and using voice awakening model | |
CN109064532B (en) | Automatic mouth shape generating method and device for cartoon character | |
CN110688518A (en) | Rhythm point determining method, device, equipment and storage medium | |
CN106653037A (en) | Audio data processing method and device | |
CN111179915A (en) | Age identification method and device based on voice | |
CN109360551B (en) | Voice recognition method and device | |
CN108711415B (en) | Method, apparatus and storage medium for correcting time delay between accompaniment and dry sound | |
CN104978961B (en) | A kind of audio-frequency processing method, device and terminal | |
CN111210850B (en) | Lyric alignment method and related product | |
WO2023116122A1 (en) | Subtitle generation method, electronic device, and computer-readable storage medium | |
CN109979434A (en) | The test method of local voice mould set product acoustical behavior | |
CN114007064B (en) | Special effect synchronous evaluation method, device, equipment and storage medium | |
CN114374885B (en) | Video key fragment determining method and device, electronic equipment and readable storage medium | |
CN113255348B (en) | Lyric segmentation method, device, equipment and medium | |
US20210407483A1 (en) | Computer-based systems, devices, and methods for generating musical compositions that are synchronized to video | |
CN104078063A (en) | Audio file playing method and system | |
CN105320641B (en) | Text verification method and user terminal | |
CN110503991A (en) | Voice broadcast method, device, electronic equipment and storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |