Kunlun Wanwei Group announced today that its music generation model "Heavenly Craftsmanship SkyMusic"Free invitation beta testing will start today. This round of invitation beta testing will offer 1,000 free places to industry media, experts, and interested music practitioners. The official version will be free to the public along with "Tiangong 3.0" on April 17.
"SkyMusic" claims to be the only publicly available AI generates large modelsSkyMusic adopts a Sora-like model architecture in the field of music and audio. Large-scale Transformer is responsible for composing music to learn the contextual dependencies of Music Patches and achieve music controllability. Diffusion Transformer is responsible for singing, and through LDM, Music Patches are restored to high-quality audio, enabling it to support the generation of 80-second 44100Hz sampling rate two-channel stereo songs.
It is reported that SkyMusic has five major features:High-quality AI music, generate 80 seconds of 44100Hz sampling rate two-channel stereo AI song,The human voice is "indistinguishable from the real thing", the Chinese level is extremely good, the pronunciation is clear without any abnormal sound,Lyrics paragraph control, the generated songs can clearly distinguish the emotional changes in different lyrics paragraphs,Various music styles, supports rap/folk/funk/ancient style/electronic, etc.Music Intelligent Expression, you can learn a variety of singing techniques such as vibrato, opera, chant, duet, automatic harmony, etc.
It is worth noting that a "music version of ChatGPT" tool, Suno V3,After the user specifies the lyrics, it can generate songs of different styles and genres, and even specify the timbre of the AI singer..