NVIDIA AI Unveils Audio-SDS: Revolutionizing Sound Effect Creation and Multi-Task Audio Processing
2025-05-12 / Read about 0 minute
Author:小编   

NVIDIA's AI research team, as reported by AIbase, has introduced a groundbreaking technology known as Audio-SDS. This advanced technology incorporates Score Distillation Sampling (SDS) into text-conditional audio diffusion models, dramatically boosting the capabilities for sound effect generation, sound source separation, and multi-task audio processing. This pioneering achievement has garnered substantial interest and discourse across both academic and industry circles.