ASRT_SpeechRecognition icon indicating copy to clipboard operation
ASRT_SpeechRecognition copied to clipboard

FileNotFoundError: [Errno 2] No such file or directory:ST-CMDS-20170001_1-OS

Open Tinrry opened this issue 4 years ago • 2 comments

I download ST-CMDS-20170001_1-OS.tar.gz, unzip in the dataset, when i run python train_mspeech.py, will raise FileNotFoundError: [Errno 2] No such file or directory: 'dataset/ST-CMDS-20170001_1-OS/20170001P00354A0118.wav' and i search the dataset file ,i found the file actually not exist in ST-CMDS-20170001_1-OS/ directory, there exist 20170001P00354A*.wav files. I think they maybe the index number mismatch. How can i solve this problem., my i modify the readdata24.py? or The dataset has been tamper?

Tinrry avatar Sep 09 '20 09:09 Tinrry

can you understand Chinese?

chinese: 我这边没有这个问题,你把代码clone后,下载tar.gz 压缩包,解压后,的目录搜一下,我能找到。并且我训练过程ok,除了后面出现了内存消漏。

我怀疑是你文件制定位置不对, 我的 解压后放在: D:\ASR_project\dataset\ST-CMDS-20170001_1-OS

下面直接就是 音频文件列表。
然后 train_mspeech .py 代码 修改如下:

system_type = plat.system() # 由于不同的系统的文件路径表示不一样,需要进行判断 if(system_type == 'Windows'): datapath = 'D:\ASR_project\dataset' modelpath = modelpath + '\' elif(system_type == 'Linux'): datapath = 'dataset' modelpath = modelpath + '/'

laoyin avatar Nov 10 '20 09:11 laoyin

我出现了同样的情况,然后我发现是因为数据本身少了几个,重新下载数据,解压,能够运行。

zh0819 avatar Jun 29 '21 06:06 zh0819