chance

The loss of S and C optimization is different

1

The paper mentions that S and C optimization have the same loss, but in the code, dice loss is added to optimize S, and after I try to get rid...

I try to use the KL loss in Mask R-CNN, though the loss was small at first(Lreg = 0.0256), it didn't change much after training 50000 times(Lreg = 0.0223). The...

The accuracy rises slowly

4

I used to train Chinese text recognition. After training for 10w iterations, the accuracy rate is still less than 0.5，while the loss value has dropped below 0.2.What might be wrong?...

Why I use the history_prompt but the result also different?

3

I use the history_prompt while set the output_full=True, and than put the same input text , but the result sounds so different. So if the history_prompt only mean use the...

克隆后的音频播放时，杂音很重

1

**合成后的杂音很重怎么处理** **Env & To Reproduce[复现与环境]** encoder: pretrained_bak_5805000 ppg_extractor：24epoch.pt ppg2mel：ppg2mel.yaml ppg2melbest_loss_step_322000.pth vocoder_hifigan_24k：hifigan_24k.pt config.json 用到的模型如上所示，我运行了run.py得到合成后的音频，播放时发现杂音非常大，合成的音色和给定的音色还算接近，如何处理掉杂音问题？ 另外我从README的2.3中下载了社区提供的预训练合成器，但不知道这个模型要用在何处？只有运行toolbox时指定了这个模型，run.py不需要用到吗？ 烦请解答，谢谢！

Different feature values while using "import torch" !

1

for example ``` import audioflux output1 = audioflux.mel_spectrogram(waveform, num=80, radix2_exp=10, samplate=16000) ``` ``` import torch import audioflux output2 = audioflux.mel_spectrogram(waveform, num=80, radix2_exp=10, samplate=16000) ``` I finally found out that the...

can't tar -xvf it

Is the accuracy of the paper reproduced?

1

Thanks for reply!

Can support ARM64?

1

I need to build audioflux within an arm64-python Docker image running a Linux system. Is this supported?

PUNC模型怎么跑onnx推断

1

https://github.com/yeyupiaoling/MASR/blob/4ee83b1e71d29e602c3aa93ec90c2761abb80cc8/masr/infer_utils/punc_predictor.py#L21 这个模型要怎么转成onnx？ 而且转完之后，在3.0版本中怎么运行onnx推断呢？压缩包里缺失相应的configs.yaml、jieba_user_dict、tokens.json（这个我在2.x版本中找到了）。 我是通过如下命令转的onnx，但显然完全错误： paddle2onnx --model_dir "pun_models" --model_filename "model.pdmodel" --params_filename model.pdiparams --save_file model.onnx 然后得到的onnx模型的输入名是input_ids和token_type_ids。 应该怎么转呀？