Audio Generation Model

On the 14th of September, Stability AI has officially released the Enterprise Audio Generation Model Stable Audio 2.5, which, relative to the previous generation, has been upgraded to focus on audio detail and production speed, stating that “it takes only two seconds to create three minutes of audio. It was described that the core improvements of Stable Audio 2.5 focused on the ability to produce music, which was described as producing results that were more accompanied by the actual chorus logic and presented a full multi-part structure of foreplay, development and end. And the new model is more accurate in its understanding of the hints..

July 4 news, Ali "Tongyi big model" public number today announced that the Tongyi laboratory's first audio generation model ThinkSound is now officially open source, will break the imagination of the "silent picture" limitations. ThinkSound applies CoT (Chain-of-Thought) to audio generation for the first time, allowing AI to learn to "think through" the relationship between picture events and sound step by step, thus realizing high-fidelity, strongly synchronized spatial audio generation - not just "picture dubbing". - This enables high-fidelity, strongly synchronized spatial audio generation - not just "looking at the picture and dubbing", but actually "understanding the picture". ...

Recently, the Stability AI team launched a new open source audio generation model called Stable Audio Open. What is special about this model is that it can generate up to 47 seconds of stereo audio from text prompts with a sampling rate of up to 44.1kHz. Unlike many currently popular audio generation models, the weights of Stable Audio Open are open, which means that anyone can view, modify, and extend this model. This design concept not only promotes the progress of scientific research, but also provides developers with…