Compared to images and video, AI generation technology for audio content is relatively mature, with text-to-speech (TTS) quality approaching and in some scenarios even reaching human-level performance.