Artificial intelligence transforms audiovisual creation methods, requiring human expertise
Artificial intelligence AI is increasingly integrated into audiovisual creation processes globally, from image generation to video production, music, special effects, and digital avatars. Wang Yihan, a mid-level engineer, discussed "Audiovisual Creation Assisted by AI" at a seminar in Beijing, China, for media professionals. She highlighted that while AI tools, particularly generative AI AIGC, can perform tasks in minutes that previously took hours or days, their effective use still demands human experimentation, creativity, and judgment. For example, an image that once required one to two days for a graphic artist can now be generated by MidJourney in under a minute from a text prompt. Early AI tools, however, were imperfect; for instance, creating three comic strip panels, readable in 30 seconds, still required two people working full-time for two to three days to finalize. The evolution accelerated with video generation tools like OpenAI's Sora in 2024, which allowed content creation from text. Initial videos had flaws such as unnatural movements and inconsistencies, but advancements now enable generating 5-10 second sequences, animating photos, and controlling object movements. Tools like Pika apply special effects and generate sound, while others allow for movement control and multimodal video editing. Image generation tools like MidJourney, Whisk Google, and Jimeng ByteDance offer extensive control over style, framing, and quality, and can even describe images automatic
