qianmo
qianmo
@olivier-gillet Have you solved this problem?
same problem , initialize the weights random
same problem, the generated voice almost as same as the driving sample. And i found that the code just fine-tuning vocoder(hifigan )
same question
持续关注该问题,不知道咋解决的
持续关注,有好的解决方案吗, 用libtorch?
same problem ,have you solved it?
Framework: onnx Not Supported OP: Transpose
数据集的格式什么呢?data/plate_real.tfrecords是真实图片数据,data/plate_synthetic.tfrecords这个数据从哪里来呢