Yang Yue, a veteran radio disc jockey in Beijing who has gained a fan base through podcasting and livestreaming his shows, recently shared a video he created by using AI software.
Within 30 days, he used nine pieces of the software to create the video, which runs for 3 minutes, 43 seconds.
The footage begins with a male robotic voice saying: "This is a dilapidated garage in New York. A group of young people who love music gather here... They rehearse day in and day out so that others can hear their songs. As they are hungry, they want the opportunity to perform in a bar."
Yang used the groundbreaking chatbot ChatGPT to write the script and used other AI software to create pictures, which were transformed into video footage with the help of more AI devices. All the music featured in the video was created by AI.
"The only elements in the video not generated by AI are the sound effects, such as rain, cars and noise from daily life, which I bought from an online material database that is already widely used," Yang said.
He added that he spent about 25 days working with AI, and the process of creating the video took him about five days.
"Write down any idea or topic you have, and AI such as ChatGPT and video generators will create a video complete with clips, subtitles and background music. However, it took me a long time to work with this software, as it didn't always give me what I asked for," Yang said.
For example, when he typed in the words "a young man in red hoodie playing guitar on the street in New York", he received four pictures - three of them not in the precise shade of red he required, and the other featuring what appeared to be a young woman holding a guitar, rather than playing it.
The original script written by Yang centered on a group of young people pursuing their musical dreams in New York. Despite their initial struggles and failures, they are eventually successful.
However, the final footage features the story of a young man who makes a deal with a scientist to create a band that never stops playing. The man plays his guitar for the scientist, along with a group of robotic bandmates.
When the scientist has acquired sufficient sources of music, which can be constantly renewed and updated, he abruptly returns the young man to the street, where he starts singing and playing his guitar again. The name of the video appears at the end of the footage-Only The Soul Can Sing.
Yang said: "The script goes in a totally new direction, and I like it very much. When I saw a picture of a man looking like an evil scientist, I decided to use it, and drew inspiration from it. From start to finish, I didn't have full control over creating and editing the footage. However, I realized that all I needed to do was work with AI. On the other hand, I did have full control of the entire creative process."
It was not the first video Yang had made with the help of AI technology, but it is his longest. The footage received warm feedback from online viewers and his friends, especially those working in the music industry.
"The use of AI could push our creativity to new levels. I always love and embrace new things. Learn about AI at the very least. If you don't, you may find you're left behind," Yang said.