When AI and humans PK compose, can digital muse bring more inspiration for music creation?

Mondo Entertainment Updated on 2024-03-03

Give the AI a text command - "Melancholy Rainy Day", and after more than 1 minute, it generates 5 15-second melodies.

At the same time, composer Zhu Xuzhi sat down by the piano and improvised a piano piece during the AI's "thinking" time.

The former contains a wealth of orchestration and has a strong sense of rhythm. The latter is just a simple piano accompaniment, but it is gentle and soothing, and both of them interpret the artistic conception of "melancholy" according to their own understanding. From March to April this year, the Shanghai ** Hall will hold the "Digital Muse - 2024 ** Science and Technology Inclusion Festival" to inspire the best with digital and support the integration and innovation of ** home exploration and modern technology. At the press conference, this scene of the composing battle between humans and AI was intuitive and shocking.

Not only the Shanghai ** Hall, but also the Shanghai Chinese Orchestra officially announced that it would cooperate with the "Tianqin Lab" under Tencent ** Entertainment Group (TME) to try the cross-border integration of AI creation and traditional folk music. In today's era of ChatGPT, SORA and other artificial intelligence models, how digital technology and AI affect artistic creation, and what role humans will play in it, have become topics that people are both concerned about and somewhat unimaginable.

At the Shanghai Spring International Festival in April this year, the Shanghai Chinese Orchestra will bring the "Zero One Chinese Color" composed by AI. The performance consists of 10 pieces, inspired by the traditional Chinese 24 solar terms and their representative colors, with the participation of more than 20 musicians of the orchestra, and will use dozens of national musical instruments.

Whether it is traditional folk instruments such as pipa, ruan, flute, guzheng, or modern pop instruments such as keyboards and jazz drums, after training with rich and diverse music styles and creative rules, and then through style optimization, it will achieve a leapfrog breakthrough in directly generating high-quality input description instructions. Zhou Wenjiang, vice president of TME technology and head of Tianqin Laboratory, said.

At present, the Shanghai Chinese Orchestra has provided performance audio materials to the laboratory, which are classified according to the labels of emotion and timbre, which are used for the training of the largest model. Suona player Hu Chenyun provided some audio materials showing the suona instrument. "Suona can not only play traditional folk music such as "Hundred Birds and Phoenix", but also jazz and rock music. We don't set limits on national musical instruments, and we don't want to set limits on the creation of AI, but explore with an open attitude. ”

Also in April this year, the "Sound of Vientiane - Electronic and Digital Image" will be unveiled at the 2024 Science and Technology Inclusion Festival. This is a cross-regional exploration brought by the creators of Shanghai University and China College of Electronic Design.

At present, AIGC (artificial intelligence-generated content) relies on a large amount of learning and training in the early stage, acting as a "scholar" to summarize the laws and particularities of creation. More than half of the R&D personnel in Tianqin Laboratory have a ** background, with basic ** literacy and computer professional knowledge, in addition to creating ** itself, they are also thinking about how to make ** works emotional.

Technically, it can be done. For example, you can use data to label the best works created by humans, classify them into different emotions such as cheerful, melancholy, and anger, and then use a lot of computer learning to find patterns, imitate human emotions, and produce works with different emotions and styles. Zhou Wenjiang introduced.

In order to allow more audiences to experience the cutting-edge development achievements of the technology industry, the "Digital Muse - 2024 Technology Inclusion Festival" will continue to hold a series of workshops from March to April: "AI vs. Human Composers", "SI-FI Science Fiction** Exploration Tour", "Harmony" Automatic Musical Instrument Installation Exhibition, Sound Art and Visual Art Exhibition and other forms.

At the press conference, young composer and pianist Zhu Xuzhi staged a short scene of "AI vs. Human Composers". In fact, after getting the text instruction "Melancholy Rainy Day", Zhu Xuzhi's response was much faster than that of AI. He came to the piano and improvised a soft and soothing melody, and the AI on the big screen was still "thinking".

And which of the two products is more in line with the artistic conception of "melancholy" may depend on the audience's **aesthetics, but it is certain that the **melody made by AI is also very mature.

Now that AI can also compose, will it take away the composer's "job" in the future?

In this regard, Gao Yuejie, senior audio algorithm engineer of the audio laboratory, said, "When we as AI R&D designers create a tool, we all hope that it can help artists speed up the transformation of inspiration into reality, but this inspiration itself still irreplaceably comes from the ** family himself." ”

For Zhu Xuzhi, who has experienced the composition war, AI is both an opponent and a friend. "When I was out on the waves during the Spring Festival holiday, it was said that this AI was still deep learning at a volume of more than 200,000 songs per minute, and the speed of artificial intelligence learning was unmatched by humans, and I have always embraced this technology. I want to hear the AI create an unimaginable moment, but maybe it's also an irreplaceable part of the human composer – there will be a touching moment in the music at any time, you don't know why, but you will be moved by it. ”

The Digital Muse - 2024 Science and Technology Inclusion Festival has a total of five units, including "Infinite Fun" performance, creative promotion meeting, workshop, special presentation and ** Technology Inclusion Forum. Among them, the content of the first three units was solicited by the Shanghai ** Hall for the world's best families, and five "chief recommendation officers" including Yu Yang, deputy director of the Art and Artificial Intelligence Committee of the Chinese Society of Artificial Intelligence, were specially invited to select. Among them, the "Infinite Fun" exhibition section screened 4 works from 102 submitted works and appeared in April in turn.

From April 4th to 5th, the "Dream Space - Xu Zhihan Jazz Orchestra ** Meeting" will be led by composer and jazz guitarist Xu Zhihan 10 outstanding ** members of the Jazz Orchestra to explore the colorful colors of jazz music by integrating Western classical **, Chinese elements and folk characteristics.

From April 12th to 13th, the "Past is the Future, Digitized World Heritage 'Nanyin+'* Meeting", the "Dianshui Nanyue" band from Quanzhou will present the world's intangible cultural heritage, the "living fossil" of ancient China - Quanzhou Nanyin, to the Shanghai audience.

From April 17th to 18th, "Mange Xing - Narrative of AI ** across time and space" tells the story of a singer who has crossed the changes of Shanghai over the past century, put his soul and memory into the system of artificial intelligence, and still makes unremitting efforts to explore the meaning of ** and art.

From April 27th to 28th, the "Sound of Vientiane - Electronic and Digital Image Conference", creators from Shanghai University and China University of Electronic Design will conduct a cross-regional and cross-cultural exploration. The event will cover a variety of cultural characteristics such as folk nursery rhymes in Shanghai, guqin culture in Jiangsu, gongs and drums in Shanxi, ethnic elements in Xinjiang, etc., through the combination of electronics, digital images, live performances by performers, and AI screen iteration and other elements, showing the cultural characteristics and sound charm of various parts of China.

In addition, the "Special Section" will welcome two Asian premieres: on April 26, "Voices of Silence" – physical performance and video presentation of Schubert's art songs, produced by the Berlin ** theater company "Nicol and the Navigator" and the Brussels Arts Center, interpreting Schubert's art songs (poetry and **) through a comprehensive and modern stage performance; On April 28th, "Power and Freedom" - physical performance and video presents Beethoven's works, performed and produced by the Berlin ** theater troupe "Nicol and the Navigator" & Schwetzingen ** Festival in Germany, with the Fugue Grand and the String Quartet in F major as an index to find the traces of the composer Beethoven.

Related Pages