AI composition was a way to use artificial intelligence technology to create music. Some AI compositions used the principle of "progression chords" in impromptu accompaniment to synthesize melodies with different tones, rhythms, and sound effects. The operator could control the "progression and chords" and set the relationship between the chapters of the song. For users with zero music theory knowledge, some AI composition tools provided a large number of pre-set "progression chords". Even if they didn't know the root, type, and other improvised accompaniment knowledge, they could still create music based on their sense of sound. For example, Stable AI's Stable Audible used more than 800,000 audio files containing music, sound effects, single instrument performances, etc. to train the model. The data set was more than 19500 hours long. It could generate up to 90 seconds of music with a wide range of styles, including piano, drum sets and other instruments, ethnic percussion, hip-hop, heavy metal and other different styles. It could even generate white noise. Meta AI's MusicGen was based on the Transformer architecture. It relied on an audio clip generated by an audio prediction. The prompt words were more free. In addition to text options, it could also be supplemented with audio files. AI composition remolded the creation model of pop music, lowering the threshold of creation, so that "everyone can compose". The user could input the music style, emotion, and other "meta-tags" to quickly get the first draft and perform a second creation. At the same time, it could increase the efficiency of creation, and could try many times in a short period of time to avoid the same works. In terms of quality, the popular songs generated by AI composition would follow the law of harmony, and more electro-musical instruments would be used to maintain a strong sense of rhythm. However, there were also problems, such as the creativity of the work, the lack of emotional expression, and the fact that the various parts of the song were easy to have a "broken feeling". In addition, the AI composition software protected the copyright of music works to a certain extent, but its copyright protection function needed to be further optimized. Moreover, as the ease of use of AI composing software increased, some musicians and enthusiasts became over-reliant on it, which might dispel the seriousness of music creation. "Luo Han's Lyric Collection" is equally exciting. Everyone is welcome to click and read it!
The following are some situations related to online AI composition: ** I. Some online AI composition tools and features ** 1. ** Heavenly Sound ** - The composition principle was based on the "progression chord" principle in improvisational accompaniment. The operator had control over the "progression chord" and could set the chapter relationship of the music, such as prelude, verse, etc. For users with zero music theory knowledge, Tian Yin had prepared a large number of pre-set "progression chords" to make it convenient for users to create music based on their sense of sound. - During the creation process, the user could choose to directly start the creation or use the template. You can choose the genre, and pay attention to the BPM (beats per minute) speed filter range on the top left. The setting of the music paragraph structure is related to the length of the music. The user should choose it reasonably according to his needs or increase or decrease it after entering the console. - In the composition interface, the user could drag the progression chord group on the right into the corresponding paragraph or add it manually. If he did not know the chord characters, he could click play to judge. You can also perform operations such as splitting the BEAT, changing the key (requires a certain foundation in music theory or a good sense of music), and so on. The top score attribute was a global attribute, and it was generally not recommended to change it drastically in the middle of the composition. After finishing the composition, click "Next" in the upper right corner to render. According to the system prompt, you may need to re-select the style (such as due to speed change). The rendering will take about 1 - 3 minutes. After that, you can play it on the workbench and listen to it. If you are satisfied, you can export the music file. If you are not satisfied, you can modify the project or change the style to render again. 2. **Stable Audible (Composing AI of Stable AI)** - The model was trained with more than 800,000 audio files, including music, sound effects, and a single instrument performance. The data set was more than 19500 hours long. - It could generate up to 90 seconds of music (the normal version could generate audio within 45 seconds, and the upgraded Pro version could generate 90 seconds). It had a wide range of styles, including simple instruments (such as piano and drums), different genre styles (such as ethnic percussion, hip-hop, heavy metal), and even white noise. - After the user registered, they could use it. For example, they could request to generate music of a specific style (such as a 30-second bass solo, 112 beats, funk style), a specific instrument combination and rhythm style (such as a catchy pop dance with tropical percussion in the middle), a specific style (such as rock music), and so on. The generation process would take about one or two minutes, but the generated music might have some flaws, such as unclear timbre and missing elements in the cue words. 3. **Meta AI's MusicGen** - Based on the Transformer architecture, the audio clip was generated based on the previous audio prediction. Currently, only the Demo was released and could be experienced on HuggingFace. - When generating music, the prompt words were more free. In addition to the text options, you could also add sound files (you could drag in reference music clips or live recordings, or you could not fill in audio prompts). However, you could only generate up to 30 seconds of audio at a time. However, it was also possible to generate long audio with the help of audio prompts, but the operation would be a little troublesome. 4. **SunoAI** - It could generate a complete song based on a paragraph of text, random tapping samples, or random humming, giving everyone the opportunity to create their own songs. "Luo Han's Lyric Collection" is equally exciting. Everyone is welcome to click and read it!
There were currently some free AI composition tools. For example, Usio could compose 1000 + songs for free every month. Users only needed to enter their favorite music style, melody, and other basic requirements on its website. It could use advanced artificial intelligence algorithms to build a knowledge base using the elements extracted from a large number of music works, and then use deep learning technology to analyze the rules between the elements, so as to customize high-quality original works for users. Tiangong's Tiangong software was completely free. After installing it on his phone, he entered the "Song Writing" interface in the "Music" function. He filled in the lyrics that he had written himself or those that were generated by AI, and chose the style to generate a complete song. There was also AI mate. After opening it, he could directly click on AI to write songs. After entering his thoughts, he could click on one button to make a song. He could generate two songs at once. When entering the description, he could add music style keywords such as rock and electronic music to determine the desired style. Finally, he could click the download button on the right to save it to his phone. "Luo Han's Lyric Collection" is equally exciting. Everyone is welcome to click and read it!
The AI composition principle was mainly as follows: - In terms of music knowledge representation, it was necessary to represent and understand music. Common representation methods included MIDI files, symbolic music, audio signals, and so on. This was to generate works that complied with human music aesthetics and creation rules. - The core of the game was machine learning algorithms, such as neural networks, deep learning, reinforcement learning, etc., which trained models to create new works by learning a large amount of music data. - The music generation model was also the key. Common rule-based models, probability-based models, and generation-adversal network models could generate new music clips based on the input music knowledge and parameters. - Models such as genetic algorithms, markov chains, and mathematical music are also used. For example, mathematical algorithms are used to create mathematical music (mostly simple works), and markov chains are based on random processes and probability logic for limited control. Genetic algorithms simulated the reproduction process of species (the arrangement and combination of musical notes were written and the excellent works were automatically selected, which had the advantages of mature algorithms and simple implementation). Through music rules, the computer could formulate rules, establish a massive database, and then carry out deep learning to analyze the composition rules, structure, and other information, and then regenerate music. - Some AI compositions used the principle of "progression chords" in improvisational accompaniment to synthesize melodies with the timbre, rhythm, and sound effects of different styles. The operator had absolute control over the "progression chords" and could set the prelude, verse, chorus, bridge, interlude, and ending. "Luo Han's Lyric Collection" is equally exciting. Everyone is welcome to click and read it!
Currently, there were many situations where AI composition was free. For example, Usio was an AI music generator. During the beta testing phase, all users could enjoy 1200 free composing credits per month. The composing steps were roughly three steps: prompt words, lyrics, and generation. It could generate music of various styles and genres according to the text, but it could only generate 33 seconds of songs at a time, and there was a low probability of drawing cards. However, the songs could be expanded forward and backward through the Extension function. In addition, in AI mate, click on AI Songwriting. After entering the idea, two songs can be generated at once. When entering the description, you can add music style keywords to create the song you want. There was also the free AI voice synthesis software jointly produced by Netease Cloud Music and Xiaomi Xiao Bing, which supported AI music composition. "Songsong AI Songwriting" could also generate compositions with one click, allowing users to easily create original songs. "Luo Han's Lyric Collection" is equally exciting. Everyone is welcome to click and read it!
There were currently many AI composing software available for the Android system. For example, the AI songwriting app was rich in functions. It had the function of sharing friends with one click, and the completed works could be saved to the device and shared; It provided customized services, with pop, rock, electronic, classical and other music styles to choose from; It could capture inspiration, save melodies, lyrics clips or inspiration notes and transform them into complete works; Its intelligent creation engine used artificial intelligence algorithms to support creation from melody generation to lyrics writing, and also supported lyrics creation in multiple languages. When registering, enter the app's [My] page and click on the avatar login instructions at the top. Enter your phone number to get the Captcha. Check the agreement and click "login" to use it. After using the tutorial, you can log in and enter the software and click on the [Creation] page. You can choose "Simple Mode" or "custom mode". In custom mode, you can enter the song name and lyrics and click [Create Song] to create (you can choose to refer to the song to create). There was also the AI Songwriter King software. Enter the song description in the input box on the home page. You can choose whether it is pure music or not, then click the generate button. Wait for about one minute, and you can see the progress of the music being generated in the list on the right. After the generation, the complete work will be displayed. You can also choose the custom mode. Enter the lyrics, style, and title, and the AI will generate music according to the lyrics. It can also randomly generate lyrics, or enter a request to generate lyrics. The system will associate the lyrics according to the keywords entered. It had functions such as creative stimulation (using AI technology to stimulate creative inspiration and break through the bottleneck of creation), efficient creation (quickly generating music materials and shortening the creation cycle), and interaction learning (learning music theory and creative skills to improve music literacy). "Luo Han's Lyric Collection" is equally exciting. Everyone is welcome to click and read it!
There were a few entertainment novels that he could recommend. Among them," Chicago 1990 " was a novel that traveled through the 1990s in Chicago. The protagonist was a mixed-race teenager who was reborn into the slums of Chicago. He realized that he and his companions had no future in the lower class. In addition," From Composers to Entertainment Superstars " was an urban romance novel. It told the story of the protagonist who became a top composer after transmigrating. He used music to change the entertainment industry and became an entertainment superstar. There were also novels by other composer who traveled through time, such as " Full-time Artist " and " Composers Reborn ". These novels all unfolded the story in the form of transmigration or rebirth, describing the protagonist's experience and achievements in the field of composition.
[Toward the Future] ##I. From the perspective of the significance of reading to growth Reading was an indispensable part of growing up. When she was young, her mother's bedtime stories started her yearning for reading, such as The Daughter of the Sea in Andersen's Fairy Tales. It made people immerse themselves in it and fantasize about creating more books. At that time, the seeds of the dream of becoming a writer were planted. As he grew older, all kinds of storybooks continued to enrich his inner world, such as "The Laughing Cat Diary" and "The Mole's Moon River", which brought him many insights. When he was slightly older, famous works by writers such as Bing Xin, Lu Xun, and Defoe entered his field of vision, such as Stars in Spring Water, Morning Flowers in the Evening, and Crusoe Crusade. It was like injecting chicken soup into the soul, allowing one to understand the positive side of society. Even though he had forgotten his books in junior high school due to his study tasks, his Chinese results had improved significantly after he picked them up. This showed that reading had a profound impact on growth and knowledge reserve, and was an important cornerstone on the road to future growth. ##2. From the perspective of adapting to a new life and cherishing the present Facing the discomfort of a new environment, such as entering a new class, leaving friends, and having communication barriers with new classmates, these were new challenges faced in growing up. However, in this process, he should learn to adapt and cherish his current life. Although they might encounter some unhappiness, such as being " stabbed " by a new classmate, it was still a part of their new life. The future was made up of the present. One should not neglect the interpersonal relationships and self-growth of the present because of the uncertainty of the future. One should actively adjust their mentality to integrate into the present environment. ##3. From the perspective of dealing with the pressure of learning and discovering the beauty of life In the process of learning, they would encounter pressure such as difficult geometry studies, and the pressure would multiply as the middle school entrance examination approached. However, in addition to the intense study, there were still beautiful things waiting to be discovered in life. It was like waiting for the epiphyllum to bloom in the courtyard. Although the process was short, the beauty when it bloomed was shocking. This inspires us to study hard for the future and deal with the pressure. At the same time, we can't ignore the small but beautiful things around us. We must learn to be balanced on the path of pursuing the future. We must know how to look for the beautiful moments in life under pressure and walk into the future with a better attitude. " Her Shenzhen-Writing the Future " is equally exciting. Everyone is welcome to click and read it!
" Marriage Composing, Divorce Composing " was an online novel. The following was the rough outline of the storyline: The protagonist, Lin Feng, was a young composer. He met the heroine, Monica, by chance and fell in love at first sight. After they got together, Lin Feng began to show Monica his musical talent, and Monica gradually fell in love with his music. However, Lin Feng and Monica's relationship did not go smoothly. Lin Feng's family did not agree with Monica's identity. They hoped that Lin Feng would marry a girl of equal social status. In addition, Lin Feng had encountered some setbacks in his work, and he began to feel tired and dissatisfied. In the end, Lin Feng decided to propose to Monica, but Monica did not agree. The two of them argued and Monica eventually left Lin Feng. After that, Lin Feng began to reflect on his feelings and began to pursue Monica again. After some hard work, Lin Feng and Monica finally got together, but their marriage was not perfect. In the days that followed, Lin Feng began to compose music again and made some achievements in music. Monica also started a new life, but she was still dissatisfied with Lin Feng's past. The whole story is full of music and emotional elements. The readers can feel the strong romantic atmosphere and deep emotional resonance.
Dumpling Creation was a well-known Chinese online novel platform with a large number of online novels, providing a convenient writing and reading experience. Dumpling's creation focused on the quality of creation and provided readers with a high-quality reading experience, so it was favored by many readers. Of course, everyone's creative preferences and tastes were different. Whether they liked Tangyuan's creation or not, they still had to try it out themselves to know. However, Dumpling's works had a certain influence in the field of Chinese online novels, and it was worth a try.