This is a modified version of NVIDIA's TalkNet. It is a controllable network that can be used for both CPU and GPU inference.
This is a modified version of NVIDIA's TalkNet. It is a controllable network that can be used for both CPU and GPU inference.