-
Notifications
You must be signed in to change notification settings - Fork 27
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
最新の Linux バージョンでのインストール #84
Comments
I am getting much further with it now. I learned about anaconda virtual environment for cuda apps which I found Extremely helpful for using old toolkit, python, gcc, etc. I was able to do extract acoustics on my source wav file. I guess it does not use GPU for that step. Is that right? When doing extract acoustics for my target it seemed to use a lot of cpu and ram but then became idle. I let it run overnight about to see if it completed. What is the best format for the input wav files? I'm currently using 32-bit signed PCM at 24000hz sample rate. I've been keeping them as one long 15 minute audio clip. Is it better to divide into multiple small clips? |
I'm sorry to intru。Can you train normally。I have the following problems. If you have the same problem and tell me the solution, thank you very much. environment:wsl2-ubuntu20.4,cuda10.1 ,python3.6.6 Exception in main training loop: division by zero Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: Original exception was: |
To @SinisterSpatula :
Sorry for a late answer, the code author has mentioned on other sites that splitting your audio files to <=15 second segments is helpful. To @ms903x1 :
I still haven't been able to get train.py to run properly, but with these i have made it further than ever before. I will update if i discover anything new. Hopefully i could help. |
Yes I did end up dividing the audio into many small clips and that worked better for me. The problem with this model is that it requires a vast amount of parallel audio clips. So far the most I have tried is one hour and a half worth of clips and the quality still wasn't what I hoped for. I'm sure with more clips it will sound better but I dont have the time a patience require to build the vast parallel data set. |
私の翻訳をお許しください。 興味深いプロジェクトをインストールすることに非常に興味がありますが、2 年間で多くのことが変わりました。 ubunto 20 と cuda 11、465 ドライバー、および Titan X Pascal カードで動作しますか? cuda 10.1 および 418 ドライバーと 1060 カードが必要ですか? tensorflow 2.5.0 または 2.0 のみで動作しますか? python 3.6 と GCC 7 が必要ですか、それとも python 3.8 と GCC 8+ で動作しますか? 環境を正しく取得するのに非常に苦労しています。 アドバイスをいただければ幸いです。 まったくアップデートする予定はありますか? ありがとうございました!
The text was updated successfully, but these errors were encountered: