I am trying to use encodec.cpp as a real-time speech encoding and decoding algorithm in my project. The input is 480 frames, and during inference with strided_conv_1d, it is necessary to perform additional padding when length <= max_pad, but it seems that this is not working. Do you know why?
I am trying to use encodec.cpp as a real-time speech encoding and decoding algorithm in my project. The input is 480 frames, and during inference with strided_conv_1d, it is necessary to perform additional padding when length <= max_pad, but it seems that this is not working. Do you know why?