moshengmao
moshengmao
when I do "If you want to train the model, run train.py ``` python3 train.py --data_dir ```",the program are interrupted by a bug about nccl . I can't fix it....
When I use VCTK-DEMAND dataset to test, I found the sample_rate of wavs in the VCTK-DEMAND/test/ is 48000, but evaluation.py > assert sr == 16000 so I add lines to...
hello,I met a problem when doing python evaluation,"File "pesq/cypesq.pyx", line 1, in init cypesq ImportError: numpy.core.multiarray failed to import (auto-generated because you didn't call 'numpy.import_array()' after cimporting numpy; use 'numpy._import_array'...
您好,我从github下载您的项目,下载voicebank+demand数据集,降采样到16khz,28 speakers那个,把测试集824个wav音频分别放到main/VoiceBank+DEMAND/wavs_noisy和wavs_clean文件夹下面,使用readme里面python inference.py --checkpoint_file best_ckpt/g_best命令, 然后我把项目生成的语音,计算pesq指数,python cal_metrics.py --input_test_file /MP-SENet-main/VoiceBank+DEMAND/test.txt --clean_wav_dir /MP-SENet-main/VoiceBank+DEMAND/wavs_clean --noisy_wav_dir /MP-SENet-main/generated_files, 得到的结果 inference得到的wav进行计算pesq得分: (mp-se) (base) i@node01:~/work/MP-SENet-main/cal_metrics$ python cal_metrics.py --input_test_file /MP-SENet-main/VoiceBank+DEMAND/test.txt --clean_wav_dir /MP-SENet-main/VoiceBank+DEMAND/wavs_clean --noisy_wav_dir /MP-SENet-main/generated_files 824 Working... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100%...