Text-to-Audio

EzAudio, a new transformer-based text-to-audio (T2A) diffusion model developed by researchers from Tencent AI Lab and Johns Hopkins University. EzAudio addresses key challenges in T2A generation, including generation quality, computational…