WO2022042418A1 - 音乐合成方法、装置、设备和计算机可读介质 - Google Patents

音乐合成方法、装置、设备和计算机可读介质 Download PDF

Info

Publication number
WO2022042418A1
WO2022042418A1 PCT/CN2021/113527 CN2021113527W WO2022042418A1 WO 2022042418 A1 WO2022042418 A1 WO 2022042418A1 CN 2021113527 W CN2021113527 W CN 2021113527W WO 2022042418 A1 WO2022042418 A1 WO 2022042418A1
Authority
WO
WIPO (PCT)
Prior art keywords
lyrics
word
music
template
target
Prior art date
Application number
PCT/CN2021/113527
Other languages
English (en)
French (fr)
Inventor
顾宇
Original Assignee
北京字节跳动网络技术有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 北京字节跳动网络技术有限公司 filed Critical 北京字节跳动网络技术有限公司
Publication of WO2022042418A1 publication Critical patent/WO2022042418A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/60—Information retrieval; Database structures therefor; File system structures therefor of audio data
    • G06F16/68—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/683—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
    • G06F16/685—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using automatically derived transcript of audio data, e.g. lyrics
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/60—Information retrieval; Database structures therefor; File system structures therefor of audio data
    • G06F16/68—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/686—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using information manually generated, e.g. tags, keywords, comments, title or artist information, time, location or usage information, user ratings
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/044—Recurrent networks, e.g. Hopfield networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/049—Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • 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
    • G10H2210/00—Aspects or methods of musical processing having intrinsic musical character, i.e. involving musical theory or musical parameters or relying on musical knowledge, as applied in electrophonic musical tools or instruments
    • G10H2210/021—Background music, e.g. for video sequences or elevator music
    • 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
    • G10H2250/00—Aspects of algorithms or signal processing methods without intrinsic musical character, yet specifically adapted for or used in electrophonic musical processing
    • G10H2250/541—Details of musical waveform synthesis, i.e. audio waveshape processing from individual wavetable samples, independently of their origin or of the sound they represent
    • G10H2250/571—Waveform compression, adapted for music synthesisers, sound banks or wavetables

Definitions

  • Embodiments of the present disclosure relate to the field of computer technology, and in particular, to a music synthesis method, apparatus, device, and computer-readable medium.
  • Some embodiments of the present disclosure propose methods, apparatuses, devices, and computer-readable media for music synthesis to solve the technical problems mentioned in the above background section.
  • some embodiments of the present disclosure provide a method for synthesizing music, the method comprising: synthesizing lyrics based on a currently displayed lyrics template and at least one word input by a user according to the lyrics template; based on the lyrics and a preset background
  • the melody of the music generates labels of musical features and language features; based on the labels of the above musical features and language features, a target dry sound is generated; based on the above target dry sound and the above preset background music, music is synthesized.
  • some embodiments of the present disclosure provide a music synthesis device, the device includes: a lyrics synthesis unit configured to synthesize lyrics based on a currently displayed lyrics template and at least one word input by a user according to the lyrics template; a label The generating unit is configured to generate labels based on the above-mentioned lyrics and the melody of the preset background music to generate musical features and language features; the dry sound generating unit is configured to generate target dry sounds based on the labels of the above-mentioned musical features and language features; The synthesis unit is configured to synthesize music based on the above-mentioned target dry sound and the above-mentioned preset background music.
  • some embodiments of the present disclosure provide an electronic device, comprising: one or more processors; a storage device on which one or more programs are stored, when one or more programs are stored by one or more The processor executes, causing the one or more processors to implement a method as described in any one of the implementations of the first aspect, or to implement a method as described in any of the implementations of the third aspect.
  • some embodiments of the present disclosure provide a computer-readable medium on which a computer program is stored, wherein the program, when executed by a processor, implements the method described in any of the implementation manners of the first aspect, or implements A method as described in any implementation of the third aspect.
  • One of the above embodiments of the present disclosure has the following beneficial effects: by filling in the currently displayed lyrics template, user-defined lyrics are obtained, and interaction with the user is realized.
  • the tags of musical features and language features required for synthesizing music can be generated.
  • a target dry voice with accurate pronunciation can be obtained, for example, the dry voice of Wang Leehom's voice with accurate pronunciation when singing the above lyrics.
  • the music corresponding to the above-mentioned self-defined lyrics filled in by the user can be obtained.
  • FIG. 1 is a schematic diagram of an application scenario of a music synthesis method according to some embodiments of the present disclosure
  • FIG. 3 is a flow chart of other embodiments of the music synthesis method according to the present disclosure.
  • FIG. 4 is a schematic diagram of another application scenario of the music synthesis method according to some embodiments of the present disclosure.
  • FIG. 5 is a schematic structural diagram of some embodiments of a music synthesis apparatus according to the present disclosure.
  • FIG. 6 is a schematic structural diagram of an electronic device suitable for implementing some embodiments of the present disclosure.
  • FIG. 1 is a schematic diagram 100 of an application scenario of a music synthesis method according to some embodiments of the present disclosure.
  • the user can fill in the lyrics for the lyrics template.
  • the user may fill in the lyrics according to the template created by the lyrics of "My Motherland and Me", and the filled lyrics are indicated by reference numeral 103 .
  • the computing device 101 synthesizes the at least one word indicated by reference numeral 103 and the lyric template indicated by reference numeral 102 to obtain the lyrics indicated by reference numeral 104 .
  • the representation data of the melody of the preset background music shown in reference numeral 105 and the lyrics shown in figure 104 are combined and processed to obtain labels of musical features and language features shown in reference numeral 106 .
  • the target dry sound is obtained by processing the labels of the musical features and language features indicated by the above reference numeral 106 .
  • a dry voice can be a pure human voice without music.
  • the target dry voices can be male, female, and children's voices without music.
  • the target dry sound and the background music indicated by reference numeral 108 are processed to obtain the music indicated by reference numeral 109 .
  • the resulting male dry voice can be mixed with the background music of "My Country and Me".
  • the music synthesis method may be performed by the above-mentioned computing device 101 .
  • the computing device 101 may be hardware or software.
  • computing device 101 When computing device 101 is hardware, it may be various computing devices with information processing capabilities, including but not limited to smartphones, tablet computers, e-book readers, laptop computers, desktop computers, servers, and the like.
  • the computing device 101 When the computing device 101 is software, it can be installed in the computing devices listed above. It can be implemented, for example, as multiple software or software modules for providing distributed services, or as a single software or software module. There is no specific limitation here.
  • the music synthesis method includes the following steps:
  • Step 201 Synthesize lyrics based on the currently displayed lyrics template and at least one word input by the user according to the lyrics template.
  • the execution body of the music synthesis method may be a server.
  • the currently displayed lyric template can be incomplete lyrics with some words missing.
  • words in certain positions in the lyrics of "My Motherland and Me” can be deleted to obtain the above lyrics template.
  • the "homeland” of "me and my homeland” can be deleted to get the template “me and my___”.
  • the user can fill in the word “motherland” in the above-mentioned "me and my motherland", and can fill in "car” to synthesize the lyrics "me and my car”.
  • step 201 may be performed as follows:
  • At least one target word is determined based on the length, content and rhyme of the above at least one word.
  • the above at least one word may include multiple filler words of a missing word in a lyrics template.
  • the lyric template is "Me and my ___ will never be separated for a moment. Wherever we go, ______.”
  • At least one of the above words includes: “car”, “motorcycle”, “never separated ' and 'Never stuck in traffic'.
  • at least one target word is a filler word of a group of lyrics templates selected from the above at least one word. For example, in the above example, "car” and “never traffic jam” can be selected as at least one target word. You can also select "car” and "never separate” as at least one target word.
  • the above-mentioned execution body first counts the length of each word in the above-mentioned at least one word, and detects whether the length is within the length range specified by the padding position in the lyrics template. Afterwards, candidate filler words whose lengths are within the length range specified by the padding position in the lyrics template are selected from the at least one word. And extract the rhyme of each word in the above candidate filler words. And compare the rhyme of each word in the candidate filler words. After that, the words with the same rhyme in the candidate filler words are grouped into a group. The words in each group are respectively input into the vocabulary database composed of pre-set sensitive words for content retrieval.
  • the word is determined as a sensitive word, not as a filler word, otherwise, as a filler word. If there are filler words in a group that can satisfy the filling conditions of each position of the lyric template at the same time. Select multiple candidate filler words for each missing position of the lyrics template from the group of candidate filler words, and input the multiple candidate filler words at each position into the preset health degree model to obtain the value of each candidate filler word. Fraction. Among them, the health model is used to score each word. Finally, a candidate filler word with a score greater than a preset threshold is selected from a plurality of candidate filler words in each missing position of the lyrics template, and the selected candidate filler word is determined as at least one target word.
  • the above sensitive words refer to words with reactionary, violent and anti-human content.
  • the above-mentioned rhyme dictionary is a dictionary that records rhyming words.
  • the above-mentioned associated words may be synonyms of the words to be replaced in the lyrics or words having the same attributes as the words to be replaced in the lyrics.
  • the user fills in "whiskey” in the original "Mojito" position, and the user is provided with a related word of "whiskey”.
  • the related word can be a word that rhymes in the lyrics template, and can be "Erguotou". Users can fill in "Erguotou” into the above lyrics template to get "Give me a cup of Erguotou”.
  • the second step is to synthesize lyrics based on the above-mentioned lyrics template and the above-mentioned at least one target word.
  • the execution body may synthesize the lyrics by filling the at least one target word into the position corresponding to the lyrics template.
  • Step 202 based on the above lyrics and the melody of the preset background music, generate tags of music features and language features.
  • the preset background music may refer to music used to adjust the atmosphere in the song.
  • the music features can be features such as the pitch, pitch, and tempo of the music.
  • the music feature tag is a tag containing feature information such as the pitch, pitch value, and tempo of the music.
  • the language features can be the tones of the language, the duration of the phonemes, and the pauses of the language.
  • the labels of language features may be labels that contain features such as the tone of the language, the duration of the phonemes, and the pauses of the language.
  • the background music of "Me and My Motherland” can be used as the preset background music
  • the melody of the background music of "Me and My Motherland” can be used as the melody of the preset background music.
  • the above-mentioned execution body may first preprocess the above-mentioned melody and lyrics, and fill in the above-mentioned lyrics into the position corresponding to the above-mentioned melody to generate a score. Afterwards, each melody segment included in the melody in the above-mentioned musical score is processed separately, so as to generate the target melody. After that, the above-mentioned target melody and the above-mentioned lyrics are input into the encoder. Finally, the output of the encoder is sampled to obtain labels for musical features and linguistic features.
  • the above-mentioned melody can be represented by a piece of musical notation (for example, notation, staff).
  • the above sheet music refers to the combination of sheet music and lyrics. And the lyrics can be filled in below the corresponding note.
  • the melody of the background music of "Me and My Motherland” is selected as the melody of the preset background music.
  • the lyrics "My little train and I are inseparable for a moment” are used as composite lyrics.
  • each melody segment included in the melody in the above-mentioned score is processed separately, thereby generating the target melody.
  • the above-mentioned target melody and the above-mentioned lyrics are input into the encoder of VAE (Variational Auto-Encoder, variational auto-encoder).
  • VAE Very Auto-Encoder, variational auto-encoder
  • the output of the encoder is sampled to obtain labels for musical features and linguistic features.
  • a second deep learning network can be used inside the encoder, for example, LSTM (Long Short-Term Memory Network, long
  • the execution body may be based on the pauses between phonemes in the phoneme sequence corresponding to the lyrics, the content of the lyrics, the tone of each character in the lyrics, and the notes of the background music melody , to generate labels for musical features and linguistic features.
  • pass lyrics into a deep learning network e.g., LSTM.
  • the pauses between phonemes in the phoneme sequence of the predicted lyrics and the tone of each word in the lyrics, as well as the content of the lyrics and the notes of the background music melody are input into the encoder of the VAE to generate labels for the music features and the language features.
  • Step 203 based on the tags of the above-mentioned musical features and language features, generate a target dry voice.
  • the target dry sound is a sound with a specific type of timbre.
  • sounds made by natural persons such as male, female and child voices.
  • the above-mentioned executive body may input the labels of the above-mentioned musical features and language features into a duration model and an acoustic model, for example, a hidden Markov model and an LSTM. And the characteristic parameters of the speech signal obtained by the above model are transmitted to the vocoder to synthesize the target dry sound.
  • a duration model and an acoustic model for example, a hidden Markov model and an LSTM.
  • Step 204 synthesizing music based on the target dry sound and the preset background music.
  • the above-mentioned execution body may perform audio processing on the above-mentioned target dry sound and the above-mentioned preset background music, and convert them into MIDI (Musical Instrument Digital Interface, musical instrument digital interface) or other audio file formats.
  • MIDI Musical Instrument Digital Interface, musical instrument digital interface
  • the above-mentioned execution body may perform at least one of the following processing on the above-mentioned target dry sound and the above-mentioned preset background music: sound modification, duration adjustment, and sound mixing.
  • the user can fill in the lyrics according to the lyrics template to obtain the lyrics required for generating music.
  • the above-mentioned lyrics and melody can be processed to obtain the dry voice of male, female or child voices required by the user.
  • Combine the dry voice with the background music for example, mix the male dry voice and the background music of the song "Me and My Motherland", perform operations such as mixing, repairing or adjusting the sound duration to obtain synthetic music.
  • Users can customize lyrics and participate in music synthesis. On the basis of the synthesized music, the user can use the above-mentioned music to make a soundtrack or make a video.
  • the music synthesis method includes the following steps:
  • Step 301 Determine whether at least one word satisfies a preset condition.
  • the preset condition may be a condition for judging the length of the above at least one word, a condition for whether it is a sensitive word, a condition for whether it rhymes, and the like.
  • not a sensitive word means that each word in the above at least one word does not have reactionary, violent or anti-human content.
  • Rhyming means that each of the at least one of the above words has the same rhyme.
  • the above-mentioned at least one word may be composed of a filler word in the missing position of each lyric in the lyric template.
  • the above-mentioned execution body may match the preset length condition of the filler word at each missing lyric position in the lyric template with the length of each word in the above-mentioned at least one word.
  • a word that satisfies the preset length condition of the filling word at the position can be found in the at least one word, and it is determined that the at least one word satisfies the preset condition. On the contrary, it is determined that the above-mentioned at least one word does not satisfy the preset condition.
  • the execution body may extract the rhyme of each word in the at least one word, and compare the rhyme of each word. In response to the rhyme of each of the above words being the same, it is determined that the above-mentioned one less word satisfies the preset condition. On the contrary, it is determined that the above-mentioned at least one word does not satisfy the preset condition.
  • the above-mentioned execution body may determine whether each word in the above-mentioned at least one word is a sensitive word.
  • each word in the above at least one word may be input into a preset database composed of sensitive words for retrieval.
  • the at least one word has a sensitive word, and it is determined that the at least one word does not meet the preset condition.
  • the above method further includes: in response to not meeting a preset condition, outputting prompt information and at least one candidate replacement word, where the candidate replacement word is related to a word that does not meet the preset condition and in response to receiving the target word selected by the user from the at least one candidate replacement word, modifying the at least one word based on the target word.
  • the above-mentioned words associated with words that do not meet the preset conditions may be synonyms or words with the same attribute.
  • a prompt message is returned to the user, for example, "the length is not satisfied, please modify the input word” or A synonymous and rhyming alternative to "little motorcycle” for the user to choose from.
  • the user modifies "small motorcycle” to "motorcycle”. After that, it is determined whether the modified at least one word satisfies the preset condition.
  • Step 302 in response to satisfying the preset condition, synthesize the lyrics based on the currently displayed lyrics template and at least one word input by the user according to the aforementioned lyrics template.
  • the length is judged and the specified length is satisfied.
  • the content is positive and healthy, and as judged by rhyme, it rhymes in the lyrics template.
  • Step 303 based on the above-mentioned lyrics and the melody of the preset background music, generate tags of musical features and language features.
  • Step 304 based on the tags of the above musical features and language features, generate a target dry voice.
  • Step 305 synthesizing music based on the target dry sound and the preset background music.
  • steps 303 , 304 and 305 for the specific implementation of steps 303 , 304 and 305 and the technical effects brought about by them, reference may be made to steps 202 , 203 and 204 in the embodiment corresponding to FIG. 2 , and details are not repeated here.
  • One of the above embodiments of the present disclosure has the following beneficial effects: by setting preset conditions, the length of the words input in the lyrics template can be made reasonable, the lyrics can be made to rhyme, and the content of the lyrics can be guaranteed to be positive and healthy. It can make the synthesized music active and healthy, and improve the quality of music melody and lyrics. It can assist the user to fill in the lyrics and help the user to complete the music synthesis.
  • FIG. 4 is a schematic diagram of another application scenario of the music synthesis method according to some embodiments of the present disclosure.
  • the length indicated by the reference numeral 402 is judged, and the specified length range is satisfied.
  • the rhyme is rhymed in the lyrics template by the sensitive word judgment indicated by reference numeral 403 and the rhyme judgment indicated by reference numeral 404 .
  • the lyrics indicated by reference numeral 407 are synthesized by combining "little motorcycle” and “never traffic jam” and the lyrics template indicated by reference numeral 406 .
  • reference numeral 400 As indicated by reference numeral 400 .
  • the "little motorcycle” and “never jam in traffic” input by the user do not satisfy the length judgment shown in reference numeral 402, the sensitive word judgment shown in reference numeral 403, or the rhyming judgment shown in reference numeral 404, then return to the user with the attached file.
  • the user inputs the selected target word.
  • the above at least one word is modified based on the above target word.
  • the present disclosure provides some embodiments of a music synthesis apparatus, these apparatus embodiments correspond to those method embodiments shown in FIG. 2 , the apparatus specifically Can be used in various electronic devices.
  • the music synthesis apparatus 500 of some embodiments includes: a lyrics synthesis unit 501 , a label generation unit 502 , a dry voice generation unit 503 and a music synthesis unit 504 .
  • the lyrics synthesis unit 501 is configured to synthesize lyrics based on the currently displayed lyrics template and at least one word input by the user according to the above-mentioned lyrics template;
  • the label generation unit 502 is configured to be based on the above-mentioned lyrics and the melody of the preset background music, Generate labels of musical features and language features;
  • dry sound generation unit 503 is configured to generate target dry sounds based on the above-mentioned labels of musical features and language features;
  • music synthesis unit 504 is configured to be based on above-mentioned target dry sounds and above-mentioned presets Background music, synthetic music.
  • the lyrics synthesis unit 501 is further configured to: determine at least one target word based on the length, content and rhyme of the at least one word; synthesize the at least one target word based on the lyrics template and the at least one target word lyrics.
  • the lyrics synthesis unit 501 is further configured to: based on a preset rhyme dictionary, determine related words of multiple rhymes corresponding to the at least one word, and use the related words as the at least one target word.
  • the lyrics synthesis unit 501 is further configured to: determine whether the above at least one word satisfies a preset condition; in response to satisfying the preset condition, based on the currently displayed lyrics template and the user according to the above lyrics template Enter at least one word to synthesize lyrics.
  • the lyrics synthesis unit 501 is further configured to: in response to not meeting the preset condition, output prompt information and at least one candidate replacement word, the candidate replacement word is the same as that which does not meet the preset condition word-related and rhyming words; in response to receiving a target word selected by the user among the at least one candidate replacement word, the at least one word is modified based on the target word.
  • the label generating unit 502 is further configured to: based on the pauses between phonemes in the phoneme sequence corresponding to the lyrics, the content of the lyrics, the tone of each word in the lyrics, and the background music The notes of the melody generate labels for the aforementioned musical features and the aforementioned linguistic features.
  • the music synthesis unit 504 is further configured to: perform at least one of the following processing on the above-mentioned target dry sound and the above-mentioned preset background music: sound modification, duration adjustment and sound mixing, to obtain the above-mentioned music .
  • the units recorded in the apparatus 500 correspond to the respective steps in the method described with reference to FIG. 2 . Therefore, the operations, features and beneficial effects described above with respect to the method are also applicable to the apparatus 500 and the units included therein, and details are not described herein again.
  • an electronic device 600 may include a processing device (eg, a central processing unit, a graphics processor, etc.) 601 that may be loaded into random access according to a program stored in a read only memory (ROM) 602 or from a storage device 608 Various appropriate actions and processes are executed by the programs in the memory (RAM) 603 . In the RAM 603, various programs and data required for the operation of the electronic device 600 are also stored.
  • the processing device 601, the ROM 602, and the RAM 603 are connected to each other through a bus 604.
  • An input/output (I/O) interface 605 is also connected to bus 604 .
  • I/O interface 605 input devices 606 including, for example, a touch screen, touchpad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; including, for example, a liquid crystal display (LCD), speakers, vibration An output device 607 of a computer, etc.; a storage device 608 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 609.
  • Communication means 609 may allow electronic device 600 to communicate wirelessly or by wire with other devices to exchange data. While FIG. 6 shows electronic device 600 having various means, it should be understood that not all of the illustrated means are required to be implemented or provided. More or fewer devices may alternatively be implemented or provided. Each block shown in FIG. 6 may represent one device, or may represent multiple devices as required.
  • the processes described above with reference to the flowcharts may be implemented as computer software programs.
  • some embodiments of the present disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the method illustrated in the flowchart.
  • the computer program may be downloaded and installed from the network via the communication device 609, or from the storage device 608, or from the ROM 602.
  • the processing device 601 When the computer program is executed by the processing device 601, the above-mentioned functions defined in the methods of some embodiments of the present disclosure are performed.
  • the computer-readable medium described in some embodiments of the present disclosure may be a computer-readable signal medium or a computer-readable storage medium, or any combination of the above two.
  • the computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus or device, or a combination of any of the above. More specific examples of computer readable storage media may include, but are not limited to, electrical connections with one or more wires, portable computer disks, hard disks, random access memory (RAM), read only memory (ROM), erasable Programmable read only memory (EPROM or flash memory), fiber optics, portable compact disk read only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.
  • a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.
  • a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code therein. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing.
  • a computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device .
  • Program code embodied on a computer readable medium may be transmitted using any suitable medium including, but not limited to, electrical wire, optical fiber cable, RF (radio frequency), etc., or any suitable combination of the foregoing.
  • the client and server can use any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol) to communicate, and can communicate with digital data in any form or medium Communication (eg, a communication network) interconnects.
  • HTTP HyperText Transfer Protocol
  • Examples of communication networks include local area networks (“LAN”), wide area networks (“WAN”), the Internet (eg, the Internet), and peer-to-peer networks (eg, ad hoc peer-to-peer networks), as well as any currently known or future development network of.
  • the above-mentioned computer-readable medium may be included in the above-mentioned electronic device; or may exist alone without being assembled into the electronic device.
  • the above-mentioned computer-readable medium carries one or more programs, and when the above-mentioned one or more programs are executed by the electronic device, the electronic device: based on the currently displayed lyrics template and at least one word input by the user according to the above-mentioned lyrics template, Synthesize lyrics; based on the above-mentioned lyrics and the melody of the preset background music, generate labels of musical features and language features; based on the labels of the above-mentioned musical features and language features, generate target dry sounds; Synthesized music.
  • Computer program code for carrying out operations of some embodiments of the present disclosure may be written in one or more programming languages, including object-oriented programming languages—such as Java, Smalltalk, C++, or a combination thereof , as well as conventional procedural programming languages - such as "C" or similar programming languages.
  • the program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer, or entirely on the remote computer or server.
  • the remote computer may be connected to the user's computer through any kind of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (eg, using an Internet service provider to via Internet connection).
  • LAN local area network
  • WAN wide area network
  • each block in the flowchart or block diagrams may represent a module, segment, or portion of code that contains one or more logical functions for implementing the specified functions executable instructions.
  • the functions noted in the blocks may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved.
  • each block of the block diagrams and/or flowchart illustrations, and combinations of blocks in the block diagrams and/or flowchart illustrations can be implemented in dedicated hardware-based systems that perform the specified functions or operations , or can be implemented in a combination of dedicated hardware and computer instructions.
  • the units described in some embodiments of the present disclosure may be implemented by means of software, and may also be implemented by means of hardware.
  • the described unit can also be provided in the processor, for example, it can be described as: a processor includes a lyrics synthesis unit, a label generation unit, a dry sound generation unit and a music synthesis unit.
  • a processor includes a lyrics synthesis unit, a label generation unit, a dry sound generation unit and a music synthesis unit.
  • the names of these units do not constitute a limitation of the unit itself in some cases, for example, the lyrics synthesis unit may also be described as "a unit for synthesizing lyrics based on a lyrics template and at least one input word".
  • exemplary types of hardware logic components include: Field Programmable Gate Arrays (FPGAs), Application Specific Integrated Circuits (ASICs), Application Specific Standard Products (ASSPs), Systems on Chips (SOCs), Complex Programmable Logical Devices (CPLDs) and more.
  • FPGAs Field Programmable Gate Arrays
  • ASICs Application Specific Integrated Circuits
  • ASSPs Application Specific Standard Products
  • SOCs Systems on Chips
  • CPLDs Complex Programmable Logical Devices
  • a method for synthesizing music comprising: synthesizing lyrics based on a currently displayed lyrics template and at least one word input by a user according to the lyrics template; based on the lyrics and preset background music Based on the melody of the music feature and the language feature, the label of the music feature and the language feature is generated; based on the label of the above music feature and the language feature, the target dry sound is generated; based on the above target dry sound and the above preset background music, music is synthesized.
  • synthesizing lyrics based on the currently displayed lyrics template and at least one word input by the user according to the lyrics template includes: determining at least one word based on the length, content and rhyme of the at least one word Target word; based on the above-mentioned lyric template and the above-mentioned at least one target word, synthesize the lyrics.
  • determining at least one target word based on the length, content, and rhyme of the at least one word includes: determining, based on a preset rhyme dictionary, related words of multiple rhymes corresponding to the at least one word , taking the above-mentioned related word as the above-mentioned at least one target word.
  • synthesizing the lyrics based on the currently displayed lyrics template and at least one word input by the user according to the lyrics template includes: determining whether the at least one word satisfies a preset condition; Assuming conditions, based on the currently displayed lyric template and at least one word input by the user according to the above lyric template, the lyrics are synthesized.
  • the above-mentioned synthesizing lyrics based on the currently displayed lyrics template and at least one word input by the user according to the above-mentioned lyrics template further includes: in response to not meeting the preset conditions, outputting prompt information and at least one Candidate replacement words, the above-mentioned candidate replacement words are words that are associated with and rhyme with words that do not meet the preset conditions; in response to receiving the target word selected by the user in the above-mentioned at least one candidate replacement word, the above-mentioned at least one Words are modified.
  • generating tags of music features and language features based on the lyrics and the melody of the preset background music including: based on the pauses between phonemes in the phoneme sequence corresponding to the lyrics, the lyrics
  • the content of the above-mentioned lyrics, the tone of each word in the above-mentioned lyrics, and the notes of the background music melody generate the labels of the above-mentioned musical characteristics and the above-mentioned language characteristics.
  • synthesizing music based on the target dry sound and the preset background music includes: performing at least one of the following processes on the target dry sound and the preset background music: sound modification, Time adjustment and mixing to get the above music.
  • a music synthesis apparatus includes: a lyrics synthesis unit configured to synthesize lyrics based on a currently displayed lyrics template and at least one word input by a user according to the lyrics template; a tag generation unit configured to into a melody based on the above-mentioned lyrics and preset background music, to generate labels of musical features and language features; a dry sound generation unit, configured to generate a target dry sound based on the labels of the above-mentioned musical features and language features; a music synthesis unit, configured Based on the above-mentioned target dry sound and the above-mentioned preset background music, music is synthesized.
  • the lyrics synthesis unit is further configured to: determine at least one target word based on the length, content and rhyme of the at least one word; synthesize lyrics based on the lyrics template and the at least one target word .
  • the lyrics synthesis unit is further configured to: based on a preset rhyme dictionary, determine a plurality of associated words of rhymes corresponding to the at least one word, and use the associated word as the at least one target word.
  • the lyric synthesis unit is further configured to: determine whether the at least one word satisfies a preset condition; in response to satisfying the preset condition, based on the currently displayed lyric template and the user input according to the lyric template of at least one word, synthesized lyrics.
  • the lyrics synthesis unit is further configured to: in response to not meeting the preset condition, output prompt information and at least one candidate replacement word, where the candidate replacement word is a word that does not meet the preset condition an associated and rhyming word; in response to receiving a target word selected by the user among the at least one candidate replacement word, modifying the at least one word based on the target word.
  • the tag generation unit is further configured to: based on the pauses between phonemes in the phoneme sequence corresponding to the lyrics, the content of the lyrics, the tone of each word in the lyrics, and the background music melody to generate labels for the above musical features and the above language features.
  • the music synthesis unit is further configured to: perform at least one of the following processing on the above-mentioned target dry sound and the above-mentioned preset background music: sound modification, duration adjustment and sound mixing to obtain the above-mentioned music.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Biomedical Technology (AREA)
  • Software Systems (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Computational Linguistics (AREA)
  • Biophysics (AREA)
  • Mathematical Physics (AREA)
  • Evolutionary Computation (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Multimedia (AREA)
  • Library & Information Science (AREA)
  • Databases & Information Systems (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Auxiliary Devices For Music (AREA)

Abstract

本公开的实施例公开了一种音乐合成方法、装置、电子设备和计算机可读介质。该方法的一具体实施方式包括：基于当前展示的歌词模板和用户根据上述歌词模板输入的至少一个词，合成歌词；基于上述歌词和预设背景音乐的旋律，生成音乐特征和语言特征的标签；基于上述音乐特征和语言特征的标签，生成目标干声；基于上述目标干声和上述预设背景音乐，合成音乐。该实施方式可以实现用户参与歌词填写，根据用户填写的歌词生成音乐。

Description

音乐合成方法、装置、设备和计算机可读介质
相关申请的交叉引用
本申请基于申请号为202010880050.9、申请日为2020年08月27日，名称为“音乐合成方法、装置、设备和计算机可读介质”的中国专利申请提出，并要求该中国专利申请的优先权，该中国专利申请的全部内容在此引入本申请作为参考。
技术领域
本公开的实施例涉及计算机技术领域，具体涉及音乐合成方法、装置、设备和计算机可读介质。
背景技术
随着计算机技术的发展和手机、电脑等电子设备的普及，各种视频音频技术出现在人们的生活当中。在终端设备的使用过程中，用户体验感越来越重要，适当参与视频音频合成可以提升用户兴趣。在对视频音频的处理上，缺少用户的参与会导致用户的体验感不佳。
发明内容
本公开的内容部分用于以简要的形式介绍构思，这些构思将在后面的具体实施方式部分被详细描述。本公开的内容部分并不旨在标识要求保护的技术方案的关键特征或必要特征，也不旨在用于限制所要求的保护的技术方案的范围。
本公开的一些实施例提出了用于音乐合成方法、装置、设备和计算机可读介质，来解决以上背景技术部分提到的技术问题。
第一方面，本公开的一些实施例提供了一种音乐合成方法，该方法包括：基于当前展示的歌词模板和用户根据上述歌词模板输入的至少一个词，合成歌词；基于上述歌词和预设背景音乐的旋律，生成音乐特征和语言特征的标签；基于上述音乐特征和语言特征的标签，生成目标干声；基 于上述目标干声和上述预设背景音乐，合成音乐。
第二方面，本公开的一些实施例提供了一种音乐合成装置，装置包括：歌词合成单元，被配置成基于当前展示的歌词模板和用户根据上述歌词模板输入的至少一个词，合成歌词；标签生成单元，被配置成基于上述歌词和预设背景音乐的旋律，生成音乐特征和语言特征的标签；干声生成单元，被配置成基于上述音乐特征和语言特征的标签，生成目标干声；音乐合成单元，被配置成基于上述目标干声和上述预设背景音乐，合成音乐。
第三方面，本公开的一些实施例提供了一种电子设备，包括：一个或多个处理器；存储装置，其上存储有一个或多个程序，当一个或多个程序被一个或多个处理器执行，使得一个或多个处理器实现如第一方面中任一实现方式描述的方法，或实现第三方面中任一实现方式描述的方法。
第四方面，本公开的一些实施例提供了一种计算机可读介质，其上存储有计算机程序，其中，程序被处理器执行时实现如第一方面中任一实现方式描述的方法，或实现如第三方面中任一实现方式描述的方法。
本公开的上述各个实施例中的一个实施例具有如下有益效果：通过对当前展示的歌词模板填词，得到用户自定义的歌词，实现了与用户之间的交互。通过对上述歌词和预设背景音乐的旋律进行处理，例如，对“我和我的祖国”歌曲背景音乐的旋律处理，可以生成合成音乐所需的音乐特征和语言特征的标签。通过上述音乐特征和语言特征的标签，可以得到发音准确的目标干声，例如，对上述歌词进行演唱的发音准确的王力宏声音的干声。基于上述目标干声和上述预设背景音乐，可以得到用户填写的上述自定义的歌词对应的音乐。
附图说明
结合附图并参考以下具体实施方式，本公开各实施例的上述和其他特征、优点及方面将变得更加明显。贯穿附图中，相同或相似的附图标记表示相同或相似的元素。应当理解附图是示意性的，原件和元素不一定按照比例绘制。
图1是根据本公开的一些实施例的音乐合成方法的一个应用场景的示意图；
图2是根据本公开的音乐合成方法的一些实施例的流程图；
图3是根据本公开的音乐合成方法的另一些实施例的流程图；
图4是根据本公开的一些实施例的音乐合成方法的另一个应用场景的示意图；
图5是根据本公开的音乐合成装置的一些实施例的结构示意图；
图6是适于用来实现本公开的一些实施例的电子设备的结构示意图。
具体实施方式
下面将参照附图更详细地描述本公开的实施例。虽然附图中显示了本公开的某些实施例，然而应当理解的是，本公开可以通过各种形式来实现，而且不应该被解释为限于这里阐述的实施例。相反，提供这些实施例是为了更加透彻和完整地理解本公开。应当理解的是，本公开的附图及实施例仅用于示例性作用，并非用于限制本公开的保护范围。
另外还需要说明的是，为了便于描述，附图中仅示出了与有关发明相关的部分。在不冲突的情况下，本公开中的实施例及实施例中的特征可以相互组合。
需要注意，本公开中提及的“第一”、“第二”等概念仅用于对不同的装置、模块或单元进行区分，并非用于限定这些装置、模块或单元所执行的功能的顺序或者相互依存关系。
需要注意，本公开中提及的“一个”、“多个”的修饰是示意性而非限制性的，本领域技术人员应当理解，除非在上下文另有明确指出，否则应该理解为“一个或多个”。
本公开实施方式中的多个装置之间所交互的消息或者信息的名称仅用于说明性的目的，而并不是用于对这些消息或信息的范围进行限制。
下面将参考附图并结合实施例来详细说明本公开。
图1是根据本公开一些实施例的音乐合成方法的一个应用场景的示意图100。
如图1所示，用户可以对歌词模板，进行歌词填写。作为示例，用户可以根据“我和我的祖国”的歌词制作的模板进行填词，填写的歌词为附图标记103所示。计算设备101将附图标记103所示至少一个词和附图标记102所示的歌词模板进行合成，得到附图标记104所示歌词。之后将附 图标记105所示的预设背景音乐的旋律的表征数据和附图104所示歌词进行组合和处理，得到附图标记106所示音乐特征和语言特征的标签。作为示例，可以将附图104所示歌词与“我和我的祖国”歌曲的主旋律音符数据进行组合，得到具有音素有长停顿和短停顿特征、音高特征和乐谱时长特征等特征的标签。将上述附图标记106所示音乐特征和语言特征的标签进行处理得到目标干声。干声可以是无音乐的纯人声。目标干声可以是无音乐的男声、女声和童声等声音。将目标干声和附图标记108所示背景音乐进行处理得到附图标记109所示音乐。作为示例，可以将得到的男声干声与“我和我的祖国”的背景音乐做混音处理。
可以理解的是，音乐合成方法可以是由上述计算设备101来执行。其中，计算设备101可以是硬件，也可以是软件。当计算设备101为硬件时，可以是具有信息处理能力的各种计算设备，包括但不限于智能手机、平板电脑、电子书阅读器、膝上型便携计算机、台式计算机、服务器等等。当计算设备101为软件时，可以安装在上述所列举的计算设备中。其可以实现成例如用来提供分布式服务的多个软件或软件模块，也可以实现成单个软件或软件模块。在此不做具体限定。
应该理解，图1中的计算设备数目仅仅是示意性的。根据实现需要，可以具有任意数目的计算设备。
继续参考图2，示出了根据本公开的音乐合成方法的一些实施例的流程200。该音乐合成方法，包括以下步骤：
步骤201，基于当前展示的歌词模板和用户根据上述歌词模板输入的至少一个词，合成歌词。
在一些实施例中，音乐合成方法的执行主体可以是服务器。当前展示的歌词模板，可以是缺失某些词的不完整歌词。作为示例，可以将“我和我的祖国”歌词中的某些位置的词删除，得到上述歌词模板。例如，可以将“我和我的祖国”的“祖国”删除，得到模板“我和我的___”。用户可以对上述“我和我的祖国”的“祖国”位置填充词，可以填充“小汽车”，合成歌词“我和我的小汽车”。
在一些实施例的一些可选的实现方式中，步骤201可以如下步骤进行：
第一步，基于上述至少一个词的长度、内容和韵脚，确定至少一个目标词。其中，上述至少一个词可以包含一个歌词模板中某个缺失词的多个填充词。作为示例，例如，歌词模板为“我和我的___，一刻也不分割。无论走到哪里，______。”上述至少一个词包括：“小汽车”、“小摩托”、“永远都不分离”及“永远都不堵车”。其中，至少一个目标词为从上述至少一个词中选出的一组歌词模板的填充词。比如，上述例子中可以选取“小汽车”及“永远也不堵车”作为至少一个目标词。也可以选取“小汽车”、“永远都不分离”作为至少一个目标词。
可选的，上述执行主体首先统计出上述至少一个词中每个词的长度，并检测该长度是否在歌词模板中填充位置规定的长度范围内。之后，从上述至少一个词中选出长度在歌词模板中填充位置规定的长度范围内的候选填充词。及将上述候选填充词中每个词的韵脚提取出来。并对候选填充词中各个词的韵脚进行比较。再之后，将候选填充词中韵脚相同的词分为一组。将各组中的词分别输入预先设定的敏感词组成的词语库中进行内容检索。响应于检索成功，则确定该词为敏感词，不作为填充词，反之，作为填充词。若某个组存在可以同时满足歌词模板各个位置的填充条件的填充词。分别从该组候选填充词中选出歌词模板每个缺失位置的多个候选填充词，并将每个位置的多个候选填充词输入到预设的健康度模型，得到每个候选填充词的分数。其中，健康度模型用于对每个词进行打分。最后，从歌词模板每个缺失位置的多个候选填充词中选取出分数大于预设的阈值的一个候选填充词，并将上述选取的候选填充词确定为至少一个目标词。其中，上述敏感词指的是具有反动、暴力和反人类等内容的词语。
作为示例，将“我和我的祖国，一刻也不可分割”，制作成歌词模板“我和我的___，一刻也不可分割”。例如，歌词模板中填充词的规定的长度范围为小于等于3。则“小火车”满足词长度在规定范围。同时，“小火车”在上述歌词模板中是押韵的。最后，通过将“小火车”输入数据库检索“小火车”不具有反动、暴力和反人类等内容。及输入健康度模型打分，在分数高于预设的阈值的情况下。在上述歌词模板中原先“祖国”位置，填写上“小火车”，合成歌词“我和我的小火车，一刻也不可分割”。
可选的，基于预设韵脚词典，确定上述至少一个词对应的多个韵脚的 关联词，将上述关联词作为至少一个目标词。其中，上述韵脚词典是一种记载押韵词的词典。上述关联词可以是歌词中需替换的词的同义词或者与歌词中需替换的词具有相同属性的词。作为示例，使用“给我一杯Mojito”制作歌词模板，得到“给我一杯___”。用户在原“Mojito”位置填写“威士忌”，这时给用户提供“威士忌”的关联词，上述关联词可以是在歌词模板中押韵的词，可以是“二锅头”。用户可以将“二锅头”填入上述歌词模板，得到“给我一杯二锅头”。
第二步，基于上述歌词模板和上述至少一个目标词，合成歌词。
在一些实施例中，上述执行主体可以通过将上述至少一个目标词，填入上述歌词模板对应的位置，合成歌词。
步骤202，基于上述歌词和预设背景音乐的旋律，生成音乐特征和语言特征的标签。
在一些实施例中，预设背景音乐可以是指在歌曲中用于调节气氛的音乐。音乐特征可以是音乐的音调、音值和节拍等特征。音乐特征的标签是包含音乐的音调、音值和节拍等特征信息的标签。语言特征可以是语言的声调、音素的时长和语言的停顿等特征。语言特征的标签可以是包含语言的声调、音素的时长和语言的停顿等特征的标签。
作为示例，可以将“我和我的祖国”的背景音乐作为预设背景音乐，“我和我的祖国”的背景音乐的旋律可以作为预设背景音乐的旋律。
在一些实施例中，上述执行主体首先可以将上述旋律和歌词做预处理，并将上述歌词填入上述旋律对应的位置，生成歌谱。之后，对上述歌谱中旋律所包括的各个旋律片段分别进行处理，从而生成目标旋律。再之后，将上述目标旋律和上述歌词输入到编码器中。最后，对编码器的输出结果进行采样，得到音乐特征和语言特征的标签。
其中，上述旋律可以用一段曲谱(例如简谱、五线谱)来表示。上述歌谱是指曲谱与歌词的组合。且歌词可以填写在相应音符的下方。
作为示例，选取“我和我的祖国”的背景音乐的旋律作为预设背景音乐的旋律。歌词“我和我的小火车，一刻也不可分割”作为合成歌词。对选取的旋律及合成歌词作预处理。并将歌词“我和我的小火车，一刻也不可分割”填入上述旋律对应的位置，生成歌谱。之后，对上述歌谱中旋律 所包括的各个旋律片段分别进行处理，从而生成目标旋律。再之后，将上述目标旋律和上述歌词输入VAE(Variational Auto-Encoder，变分自编码器)的编码器中。最后，对编码器的输出结果进行采样，得到音乐特征和语言特征的标签。其中，编码器内部可以使用第二深度学习网络，例如，LSTM(Long Short-Term Memory Network，长短期记忆网络)。
在一些实施例的一些可选的实现方式中，上述执行主体可以基于上述歌词对应的音素序列中音素之间的停顿、上述歌词的内容、上述歌词中每个字的声调和背景音乐旋律的音符，生成音乐特征和语言特征的标签。作为示例，将歌词传入深度学习网络，例如，LSTM。将预测歌词的音素序列中音素之间的停顿和歌词中每个字的声调以及上述歌词的内容和背景音乐旋律的音符，输入到VAE的编码器中，生成音乐特征和上述语言特征的标签。
步骤203，基于上述音乐特征和语言特征的标签，生成目标干声。
在一些实施例中，上述目标干声是具有特定类型音色的声音。例如，男声、女生和童声等自然人发出的声音。
在一些实施例中，上述执行主体可以将上述音乐特征和语言特征的标签输入到时长模型和声学模型，例如，隐马尔科夫模型和LSTM。并将上述模型得到的语音信号的特征参量，传入到声码器合成目标干声。
步骤204，基于上述目标干声和上述预设背景音乐，合成音乐。
在一些实施例中，上述执行主体可以将上述目标干声和上述预设背景音乐做音频处理，转化为MIDI(Musical Instrument Digital Interface，乐器数字接口)或其他音频文件格式。
在一些实施例的一些可选的实现方式中，上述执行主体可以对上述目标干声和上述预设背景音乐进行以下至少一项处理：修音、时长调整和混音。
本公开的上述各个实施例中的一个实施例具有如下有益效果：用户可以根据歌词模板进行填词，得到生成音乐所需的歌词。可以对上述歌词和旋律进行处理，得到用户需要的男声、女生或童声的干声等。将干声和背景音乐组合，例如，将男声干声和“我和我的祖国”歌曲的背景音乐，进行混音、修音或声音时长调整等操作，得到合成音乐。用户可以自定义歌 词，参与音乐合成。在得到合成音乐的基础上，用户可以使用上述音乐进行配乐或制作视频。
继续参考图3，示出了根据本公开的音乐合成方法的另一些实施例的流程300。该音乐合成方法，包括以下步骤：
步骤301，确定至少一个词是否满足预设条件。
在一些实施例中，预设条件可以是判断上述至少一个词长度的条件、是否为敏感词的条件或是否押韵的条件等。其中，不是敏感词指上述至少一个词中每个词均不具有反动、暴力、反人类的内容。押韵指的是上述至少一个词中的各个词具有相同的韵脚。
在一些实施例中，上述至少一个词可以是由歌词模板中各个歌词缺失位置的一个填充词组成。
在一些实施例中，上述执行主体可以将歌词模板中每个缺失歌词位置处填充词的预设长度条件与上述至少一个词中的每个词的长度进行匹配。响应于歌词模板中每个缺失歌词位置均可以在上述至少一个词中找到满足该位置填充词的预设长度条件的词，确定上述至少一个词满足预设条件。反之，确定上述至少一个词不满足预设条件。
在一些实施例中，上述执行主体可以将上述至少一个词中的每个词的韵脚提取出来，并对各个词的韵脚进行比较。响应于上述各个词的韵脚均相同，确定上述执少一个词是满足预设条件。反之，确定上述至少一个词不满足预设条件。
在一些实施例中，上述执行主体可以判断上述至少一个词中的每个词是否为敏感词。作为示例，可以将上述至少一个词中的每个词输入到预先设定的由敏感词组成的数据库，进行检索。响应于预先设定的由敏感词组成的数据库中存在上述至少一个词中的某个词，则上述至少一个词存在敏感词，确定上述至少一个词不满足预设条件。
在一些实施例的一些可选的实现方式中，上述方法还包括：响应于不满足预设条件，输出提示信息和至少一个候选替换词，上述候选替换词是与不满足预设条件的词相关联的词；响应于接收到用户在上述至少一个候选替换词中选取的目标词，基于上述目标词对上述至少一个词进行修改。 其中，上述与不满足预设条件的词相关联的词可以是同义词或者具有相同属性的词。
作为示例，用户输入的“小摩托”和“永远都不会堵车”不满足长度判断、健康度判断或押韵判断，则返回给用户提示信息，例如，“长度不满足，请修改输入词”或可替代“小摩托”的同义并押韵的词供用户挑选。例如，用户将“小摩托”修改为“摩托”。之后，确定修改后的至少一个词是否满足预设条件。
步骤302，响应于满足预设条件，基于当前展示的歌词模板和用户根据上述歌词模板输入的至少一个词，合成歌词。
在一些实施例中，作为示例，响应于用户要填写的词“小摩托”和“永远都不会堵车”经长度判断，满足规定长度。经健康度判断，内容积极健康，以及经押韵判断，在歌词模板中押韵。将“小摩托”和“永远都不会堵车”和歌词模板合成歌词。
步骤303，基于上述歌词和预设背景音乐的旋律，生成音乐特征和语言特征的标签。
步骤304，基于上述音乐特征和语言特征的标签，生成目标干声。
步骤305，基于上述目标干声和上述预设背景音乐，合成音乐。
在一些实施例中，步骤303、304和305的具体实现及其所带来的技术效果，可以参考图2对应的实施例中的步骤202、203和204，在此不再赘述。
本公开的上述各个实施例中的一个实施例具有如下有益效果：通过设置预设条件，可以使在歌词模板中输入的词长度合理，可以使歌词押韵，以及保证歌词内容积极健康。可以使合成的音乐积极健康，提高音乐旋律和歌词的品质。可以辅助用户填写歌词，帮助用户完成音乐的合成。
图4是根据本公开一些实施例的音乐合成方法的另一个应用场景示意图。
如附图标记400所示。响应于附图标记401所示“小摩托”和“永远都不会堵车”经附图标记402所示的长度判断，满足规定的长度范围。经附图标记403所示的敏感词判断，以及经附图标记404所示的押韵判断， 在歌词模板中押韵。将“小摩托”和“永远都不会堵车”和附图标记406所示歌词模板合成附图标记407所示歌词。作为示例，继续参考图4，如附图标记400所示。用户输入的“小摩托”和“永远都不会堵车”不满足附图标记402所示长度判断、附图标记403所示敏感词判断或附图标记404所示押韵判断，则返回给用户附图标记405所示提示信息，例如，“长度不满足，请修改输入词”或可替代“小摩托”的同义并押韵的词供用户挑选。基于提示信息及可替代的词，用户输入选取的目标词。基于上述目标词对上述至少一个词进行修改。
进一步参考图5，作为对上述各图所示方法的实现，本公开提供了一种音乐合成装置的一些实施例，这些装置实施例与图2所示的那些方法实施例相对应，该装置具体可以应用于各种电子设备中。
如图5所示，一些实施例的音乐合成装置500包括：歌词合成单元501、标签生成单元502、干声生成单元503和音乐合成单元504。其中，歌词合成单元501，被配置成基于当前展示的歌词模板和用户根据上述歌词模板输入的至少一个词，合成歌词；标签生成单元502，被配置成基于上述歌词和预设背景音乐的旋律，生成音乐特征和语言特征的标签；干声生成单元503，被配置成基于上述音乐特征和语言特征的标签，生成目标干声；音乐合成单元504，被配置成基于上述目标干声和上述预设背景音乐，合成音乐。
在一些实施例的可选实现方式中，歌词合成单元501进一步被配置成：基于上述至少一个词的长度、内容和韵脚，确定至少一个目标词；基于上述歌词模板和上述至少一个目标词，合成歌词。
在一些实施例的可选实现方式中，歌词合成单元501进一步被配置成：基于预设韵脚词典，确定上述至少一个词对应的多个韵脚的关联词，将上述关联词作为上述至少一个目标词。
在一些实施例的可选实现方式中，歌词合成单元501进一步被配置成：确定上述至少一个词是否满足预设条件；响应于满足预设条件，基于当前展示的歌词模板和用户根据上述歌词模板输入的至少一个词，合成歌词。
在一些实施例的可选实现方式中，歌词合成单元501进一步被配置成： 响应于不满足预设条件，输出提示信息和至少一个候选替换词，上述候选替换词是与不满足预设条件的词相关联且押韵的词；响应于接收到用户在上述至少一个候选替换词中选取的目标词，基于上述目标词对上述至少一个词进行修改。
在一些实施例的可选实现方式中，标签生成单元502进一步被配置成：基于上述歌词对应的音素序列中音素之间的停顿、上述歌词的内容、上述歌词中每个字的声调和背景音乐旋律的音符，生成上述音乐特征和上述语言特征的标签。
在一些实施例的可选实现方式中，音乐合成单元504进一步被配置成：对上述目标干声和上述预设背景音乐进行以下至少一项处理：修音、时长调整和混音，得到上述音乐。
可以理解的是，该装置500中记载的诸单元与参考图2描述的方法中的各个步骤相对应。由此，上文针对方法描述的操作、特征以及产生的有益效果同样适用于装置500及其中包含的单元，在此不再赘述。
如图6所示，电子设备600可以包括处理装置(例如中央处理器、图形处理器等)601，其可以根据存储在只读存储器(ROM)602中的程序或者从存储装置608加载到随机访问存储器(RAM)603中的程序而执行各种适当的动作和处理。在RAM 603中，还存储有电子设备600操作所需的各种程序和数据。处理装置601、ROM 602以及RAM 603通过总线604彼此相连。输入/输出(I/O)接口605也连接至总线604。
通常，以下装置可以连接至I/O接口605：包括例如触摸屏、触摸板、键盘、鼠标、摄像头、麦克风、加速度计、陀螺仪等的输入装置606；包括例如液晶显示器(LCD)、扬声器、振动器等的输出装置607；包括例如磁带、硬盘等的存储装置608；以及通信装置609。通信装置609可以允许电子设备600与其他设备进行无线或有线通信以交换数据。虽然图6示出了具有各种装置的电子设备600，但是应理解的是，并不要求实施或具备所有示出的装置。可以替代地实施或具备更多或更少的装置。图6中示出的每个方框可以代表一个装置，也可以根据需要代表多个装置。
特别地，根据本公开的一些实施例，上文参考流程图描述的过程可以 被实现为计算机软件程序。例如，本公开的一些实施例包括一种计算机程序产品，其包括承载在计算机可读介质上的计算机程序，该计算机程序包含用于执行流程图所示的方法的程序代码。在这样的一些实施例中，该计算机程序可以通过通信装置609从网络上被下载和安装，或者从存储装置608被安装，或者从ROM 602被安装。在该计算机程序被处理装置601执行时，执行本公开的一些实施例的方法中限定的上述功能。
需要说明的是，本公开的一些实施例所述的计算机可读介质可以是计算机可读信号介质或者计算机可读存储介质或者是上述两者的任意组合。计算机可读存储介质例如可以是——但不限于——电、磁、光、电磁、红外线、或半导体的系统、装置或器件，或者任意以上的组合。计算机可读存储介质的更具体的例子可以包括但不限于：具有一个或多个导线的电连接、便携式计算机磁盘、硬盘、随机访问存储器(RAM)、只读存储器(ROM)、可擦式可编程只读存储器(EPROM或闪存)、光纤、便携式紧凑磁盘只读存储器(CD-ROM)、光存储器件、磁存储器件、或者上述的任意合适的组合。
在本公开的一些实施例中，计算机可读存储介质可以是任何包含或存储程序的有形介质，该程序可以被指令执行系统、装置或者器件使用或者与其结合使用。而在本公开的一些实施例中，计算机可读信号介质可以包括在基带中或者作为载波一部分传播的数据信号，其中承载了计算机可读的程序代码。这种传播的数据信号可以采用多种形式，包括但不限于电磁信号、光信号或上述的任意合适的组合。计算机可读信号介质还可以是计算机可读存储介质以外的任何计算机可读介质，该计算机可读信号介质可以发送、传播或者传输用于由指令执行系统、装置或者器件使用或者与其结合使用的程序。计算机可读介质上包含的程序代码可以用任何适当的介质传输，包括但不限于：电线、光缆、RF(射频)等等，或者上述的任意合适的组合。
在一些实施方式中，客户端、服务器可以利用诸如HTTP(HyperText Transfer Protocol，超文本传输协议)之类的任何当前已知或未来研发的网络协议进行通信，并且可以与任意形式或介质的数字数据通信(例如，通信网络)互连。通信网络的示例包括局域网(“LAN”)，广域网(“WAN”)， 网际网(例如，互联网)以及端对端网络(例如，ad hoc端对端网络)，以及任何当前已知或未来研发的网络。
上述计算机可读介质可以是上述电子设备中所包含的；也可以是单独存在，而未装配入该电子设备中。上述计算机可读介质承载有一个或者多个程序，当上述一个或者多个程序被该电子设备执行时，使得该电子设备：基于当前展示的歌词模板和用户根据上述歌词模板输入的至少一个词，合成歌词；基于上述歌词和预设背景音乐的旋律，生成音乐特征和语言特征的标签；基于上述音乐特征和语言特征的标签，生成目标干声；基于上述目标干声和上述预设背景音乐，合成音乐。
可以以一种或多种程序设计语言或其组合来编写用于执行本公开的一些实施例的操作的计算机程序代码，所述程序设计语言包括面向对象的程序设计语言—诸如Java、Smalltalk、C++，还包括常规的过程式程序设计语言—诸如“C”语言或类似的程序设计语言。程序代码可以完全地在用户计算机上执行、部分地在用户计算机上执行、作为一个独立的软件包执行、部分在用户计算机上部分在远程计算机上执行、或者完全在远程计算机或服务器上执行。在涉及远程计算机的情形中，远程计算机可以通过任意种类的网络——包括局域网(LAN)或广域网(WAN)——连接到用户计算机，或者，可以连接到外部计算机(例如利用因特网服务提供商来通过因特网连接)。
附图中的流程图和框图，图示了按照本公开各种实施例的系统、方法和计算机程序产品的可能实现的体系架构、功能和操作。在这点上，流程图或框图中的每个方框可以代表一个模块、程序段、或代码的一部分，该模块、程序段、或代码的一部分包含一个或多个用于实现规定的逻辑功能的可执行指令。也应当注意，在有些作为替换的实现中，方框中所标注的功能也可以以不同于附图中所标注的顺序发生。例如，两个接连地表示的方框实际上可以基本并行地执行，它们有时也可以按相反的顺序执行，这依所涉及的功能而定。也要注意的是，框图和/或流程图中的每个方框、以及框图和/或流程图中的方框的组合，可以用执行规定的功能或操作的专用的基于硬件的系统来实现，或者可以用专用硬件与计算机指令的组合来实现。
描述于本公开的一些实施例中的单元可以通过软件的方式实现，也可以通过硬件的方式来实现。所描述的单元也可以设置在处理器中，例如，可以描述为：一种处理器包括歌词合成单元、标签生成单元、干声生成单元和音乐合成单元。其中，这些单元的名称在某种情况下并不构成对该单元本身的限定，例如，歌词合成单元还可以被描述为“基于歌词模板和输入的至少一个词，合成歌词的单元”。
本文中以上描述的功能可以至少部分地由一个或多个硬件逻辑部件来执行。例如，非限制性地，可以使用的示范类型的硬件逻辑部件包括：现场可编程门阵列(FPGA)、专用集成电路(ASIC)、专用标准产品(ASSP)、片上系统(SOC)、复杂可编程逻辑设备(CPLD)等等。
根据本公开的一个或多个实施例，提供了一种音乐合成方法，包括：基于当前展示的歌词模板和用户根据上述歌词模板输入的至少一个词，合成歌词；基于上述歌词和预设背景音乐的旋律，生成音乐特征和语言特征的标签；基于上述音乐特征和语言特征的标签，生成目标干声；基于上述目标干声和上述预设背景音乐，合成音乐。
根据本公开的一个或多个实施例，上述基于当前展示的歌词模板和用户根据上述歌词模板输入的至少一个词，合成歌词，包括：基于上述至少一个词的长度、内容和韵脚，确定至少一个目标词；基于上述歌词模板和上述至少一个目标词，合成歌词。
根据本公开的一个或多个实施例，上述基于上述至少一个词的长度、内容和韵脚，确定至少一个目标词，包括：基于预设韵脚词典，确定上述至少一个词对应的多个韵脚的关联词，将上述关联词作为上述至少一个目标词。
根据本公开的一个或多个实施例，上述基于当前展示的歌词模板和用户根据上述歌词模板输入的至少一个词，合成歌词，包括：确定上述至少一个词是否满足预设条件；响应于满足预设条件，基于当前展示的歌词模板和用户根据上述歌词模板输入的至少一个词，合成歌词。
根据本公开的一个或多个实施例，上述基于当前展示的歌词模板和用户根据上述歌词模板输入的至少一个词，合成歌词，还包括：响应于不满足预设条件，输出提示信息和至少一个候选替换词，上述候选替换词是与 不满足预设条件的词相关联且押韵的词；响应于接收到用户在上述至少一个候选替换词中选取的目标词，基于上述目标词对上述至少一个词进行修改。
根据本公开的一个或多个实施例，上述基于上述歌词和预设背景音乐的旋律，生成音乐特征和语言特征的标签，包括：基于上述歌词对应的音素序列中音素之间的停顿、上述歌词的内容、上述歌词中每个字的声调和背景音乐旋律的音符，生成上述音乐特征和上述语言特征的标签。
根据本公开的一个或多个实施例，上述基于上述目标干声和上述预设背景音乐，合成音乐，包括：对上述目标干声和上述预设背景音乐进行以下至少一项处理：修音、时长调整和混音，得到上述音乐。
根据本公开的一个或多个实施例，音乐合成装置包括：歌词合成单元，被配置成基于当前展示的歌词模板和用户根据上述歌词模板输入的至少一个词，合成歌词；标签生成单元，被配置成基于上述歌词和预设背景音乐的旋律，生成音乐特征和语言特征的标签；干声生成单元，被配置成基于上述音乐特征和语言特征的标签，生成目标干声；音乐合成单元，被配置成基于上述目标干声和上述预设背景音乐，合成音乐。
根据本公开的一个或多个实施例，歌词合成单元进一步被配置成：基于上述至少一个词的长度、内容和韵脚，确定至少一个目标词；基于上述歌词模板和上述至少一个目标词，合成歌词。
根据本公开的一个或多个实施例，歌词合成单元进一步被配置成：基于预设韵脚词典，确定上述至少一个词对应的多个韵脚的关联词，将上述关联词作为上述至少一个目标词。
根据本公开的一个或多个实施例，歌词合成单元进一步被配置成：确定上述至少一个词是否满足预设条件；响应于满足预设条件，基于当前展示的歌词模板和用户根据上述歌词模板输入的至少一个词，合成歌词。
根据本公开的一个或多个实施例，歌词合成单元进一步被配置成：响应于不满足预设条件，输出提示信息和至少一个候选替换词，上述候选替换词是与不满足预设条件的词相关联且押韵的词；响应于接收到用户在上述至少一个候选替换词中选取的目标词，基于上述目标词对上述至少一个词进行修改。
根据本公开的一个或多个实施例，标签生成单元进一步被配置成：基于上述歌词对应的音素序列中音素之间的停顿、上述歌词的内容、上述歌词中每个字的声调和背景音乐旋律的音符，生成上述音乐特征和上述语言特征的标签。
根据本公开的一个或多个实施例，音乐合成单元进一步被配置成：对上述目标干声和上述预设背景音乐进行以下至少一项处理：修音、时长调整和混音，得到上述音乐。

Claims (10)

  1. 一种音乐合成方法，包括：
    基于当前展示的歌词模板和用户根据所述歌词模板输入的至少一个词，合成歌词；
    基于所述歌词和预设背景音乐的旋律，生成音乐特征和语言特征的标签；
    基于所述音乐特征和语言特征的标签，生成目标干声；
    基于所述目标干声和所述预设背景音乐，合成音乐。
  2. 根据权利要求1所述的方法，其中，所述基于当前展示的歌词模板和用户根据所述歌词模板输入的至少一个词，合成歌词，包括：
    基于所述至少一个词的长度、内容和韵脚，确定至少一个目标词；
    基于所述歌词模板和所述至少一个目标词，合成歌词。
  3. 根据权利要求2所述的方法，其中，所述基于所述至少一个词的长度、内容和韵脚，确定至少一个目标词，包括：
    基于预设韵脚词典，确定所述至少一个词对应的多个韵脚的关联词，将所述关联词作为所述至少一个目标词。
  4. 根据权利要求1所述的方法，其中，所述基于当前展示的歌词模板和用户根据所述歌词模板输入的至少一个词，合成歌词，包括：
    确定所述至少一个词是否满足预设条件；
    响应于满足预设条件，基于当前展示的歌词模板和用户根据所述歌词模板输入的至少一个词，合成歌词。
  5. 根据权利要求4所述的方法，其中，所述基于当前展示的歌词模板和用户根据所述歌词模板输入的至少一个词，合成歌词，还包括：
    响应于不满足预设条件，输出提示信息和至少一个候选替换词，所述 候选替换词是与不满足预设条件的词相关联且押韵的词；
    响应于接收到用户在所述至少一个候选替换词中选取的目标词，基于所述目标词对所述至少一个词进行修改。
  6. 根据权利要求1所述的方法，其中，所述基于所述歌词和预设背景音乐的旋律，生成音乐特征和语言特征的标签，包括：
    基于所述歌词对应的音素序列中音素之间的停顿、所述歌词的内容、所述歌词中每个字的声调和背景音乐旋律的音符，生成所述音乐特征和所述语言特征的标签。
  7. 根据权利要求1所述的方法，其中，所述基于所述目标干声和所述预设背景音乐，合成音乐，包括：
    对所述目标干声和所述预设背景音乐进行以下至少一项处理：修音、时长调整和混音，得到所述音乐。
  8. 一种音乐合成装置，包括：
    歌词合成单元，被配置成基于当前展示的歌词模板和用户根据所述歌词模板输入的至少一个词，合成歌词；
    标签生成单元，被配置成基于所述歌词和预设背景音乐的旋律，生成音乐特征和语言特征的标签；
    干声生成单元，被配置成基于所述音乐特征和语言特征的标签，生成目标干声；
    音乐合成单元，被配置成基于所述目标干声和所述预设背景音乐，合成音乐。
  9. 一种电子设备，包括：
    一个或多个处理器；
    存储装置，其上存储有一个或多个程序，
    当所述一个或多个程序被所述一个或多个处理器执行，使得所述一个或多个处理器实现如权利要求1-7中任一所述的方法。
  10. 一种计算机可读介质，其上存储有计算机程序，其中，所述程序被处理器执行时实现如权利要求1-7中任一所述的方法。
PCT/CN2021/113527 2020-08-27 2021-08-19 音乐合成方法、装置、设备和计算机可读介质 WO2022042418A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010880050.9 2020-08-27
CN202010880050.9A CN112035699A (zh) 2020-08-27 2020-08-27 音乐合成方法、装置、设备和计算机可读介质

Publications (1)

Publication Number Publication Date
WO2022042418A1 true WO2022042418A1 (zh) 2022-03-03

Family

ID=73585837

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/113527 WO2022042418A1 (zh) 2020-08-27 2021-08-19 音乐合成方法、装置、设备和计算机可读介质

Country Status (2)

Country Link
CN (1) CN112035699A (zh)
WO (1) WO2022042418A1 (zh)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112035699A (zh) * 2020-08-27 2020-12-04 北京字节跳动网络技术有限公司 音乐合成方法、装置、设备和计算机可读介质
CN113066459B (zh) * 2021-03-24 2023-05-30 平安科技（深圳）有限公司 基于旋律的歌曲信息合成方法、装置、设备及存储介质
CN114023286A (zh) * 2021-10-15 2022-02-08 天翼爱音乐文化科技有限公司 一种说唱音乐生成方法、系统、装置与存储介质

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108877766A (zh) * 2018-07-03 2018-11-23 百度在线网络技术（北京）有限公司 歌曲合成方法、装置、设备及存储介质
CN109829482A (zh) * 2019-01-04 2019-05-31 平安科技（深圳）有限公司 歌曲训练数据处理方法、装置及计算机可读存储介质
CN110516110A (zh) * 2019-07-22 2019-11-29 平安科技（深圳）有限公司 歌曲生成方法、装置、计算机设备及存储介质
WO2020077262A1 (en) * 2018-10-11 2020-04-16 WaveAI Inc. Method and system for interactive song generation
CN111445897A (zh) * 2020-03-23 2020-07-24 北京字节跳动网络技术有限公司 歌曲生成方法、装置、可读介质及电子设备
CN111506767A (zh) * 2020-03-04 2020-08-07 北京灵动音科技有限公司 一种歌曲填词的处理方法及装置、电子设备和存储介质
CN112035699A (zh) * 2020-08-27 2020-12-04 北京字节跳动网络技术有限公司 音乐合成方法、装置、设备和计算机可读介质

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103973955B (zh) * 2013-01-28 2017-08-25 联想(北京)有限公司 一种信息处理方法及电子设备
CN110019919B (zh) * 2017-09-30 2022-07-26 腾讯科技（深圳）有限公司 一种押韵歌词的生成方法和装置
CN111402843B (zh) * 2020-03-23 2021-06-11 北京字节跳动网络技术有限公司 说唱音乐生成方法、装置、可读介质及电子设备
CN111554267A (zh) * 2020-04-23 2020-08-18 北京字节跳动网络技术有限公司 音频合成方法、装置、电子设备和计算机可读介质

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108877766A (zh) * 2018-07-03 2018-11-23 百度在线网络技术（北京）有限公司 歌曲合成方法、装置、设备及存储介质
WO2020077262A1 (en) * 2018-10-11 2020-04-16 WaveAI Inc. Method and system for interactive song generation
CN109829482A (zh) * 2019-01-04 2019-05-31 平安科技（深圳）有限公司 歌曲训练数据处理方法、装置及计算机可读存储介质
CN110516110A (zh) * 2019-07-22 2019-11-29 平安科技（深圳）有限公司 歌曲生成方法、装置、计算机设备及存储介质
CN111506767A (zh) * 2020-03-04 2020-08-07 北京灵动音科技有限公司 一种歌曲填词的处理方法及装置、电子设备和存储介质
CN111445897A (zh) * 2020-03-23 2020-07-24 北京字节跳动网络技术有限公司 歌曲生成方法、装置、可读介质及电子设备
CN112035699A (zh) * 2020-08-27 2020-12-04 北京字节跳动网络技术有限公司 音乐合成方法、装置、设备和计算机可读介质

Also Published As

Publication number Publication date
CN112035699A (zh) 2020-12-04

Similar Documents

Publication Publication Date Title
WO2022042418A1 (zh) 音乐合成方法、装置、设备和计算机可读介质
WO2021218324A1 (zh) 歌曲合成方法、装置、可读介质及电子设备
CN111899720B (zh) 用于生成音频的方法、装置、设备和介质
WO2022095743A1 (zh) 语音合成方法、装置、存储介质及电子设备
WO2022188734A1 (zh) 一种语音合成方法、装置以及可读存储介质
WO2022143058A1 (zh) 语音识别方法、装置、存储介质及电子设备
CN111402843B (zh) 说唱音乐生成方法、装置、可读介质及电子设备
CN111445892B (zh) 歌曲生成方法、装置、可读介质及电子设备
CN111402842B (zh) 用于生成音频的方法、装置、设备和介质
JP6078964B2 (ja) 音声対話システム及びプログラム
CN107516511A (zh) 意图识别和情绪的文本到语音学习系统
CN111445897B (zh) 歌曲生成方法、装置、可读介质及电子设备
WO2022156413A1 (zh) 语音风格的迁移方法、装置、可读介质和电子设备
CN111798821B (zh) 声音转换方法、装置、可读存储介质及电子设备
WO2021227707A1 (zh) 音频合成方法、装置、计算机可读介质及电子设备
CN111782576B (zh) 背景音乐的生成方法、装置、可读介质、电子设备
CN112802446B (zh) 音频合成方法及装置、电子设备和计算机可读存储介质
WO2022237665A1 (zh) 语音合成方法、装置、电子设备和存储介质
WO2022184055A1 (zh) 文章的语音播放方法、装置、设备、存储介质及程序产品
WO2022111242A1 (zh) 旋律生成方法、装置、可读介质及电子设备
WO2021212954A1 (zh) 极低资源下的特定发音人情感语音合成方法及装置
CN111161695B (zh) 歌曲生成方法和装置
WO2023160553A1 (zh) 语音合成方法、装置、计算机可读介质及电子设备
CN111354343B (zh) 语音唤醒模型的生成方法、装置和电子设备
CN111477210A (zh) 语音合成方法和装置

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: 21860256

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 1205 DATED 23/06/2023)

122 Ep: pct application non-entry in european phase

Ref document number: 21860256

Country of ref document: EP

Kind code of ref document: A1