no code implementations • 17 Jun 2022 • Bang Zeng, Hongbing Suo, Yulong Wan, Ming Li
The common target speech separation directly estimate the target source, ignoring the interrelationship between different speakers at each frame.
Action Detection Activity Detection +3