deep-voice-conversion icon indicating copy to clipboard operation
deep-voice-conversion copied to clipboard

Published 20 hours ago • verified publisher icon andabi

Voice quality not good

Open hassanShabbir1960 opened this issue 5 years ago • 8 comments

loss I trained the net2 for almost 3 days and the loss value was 0.0063382 on arctic dataset. Now when I am using the model, the quality of the sound is not good and not clearly audible. Can you please suggest what I can do to improve it ? @andabi

hassanShabbir1960 avatar Nov 25 '19 06:11 hassanShabbir1960

loss I trained the net2 for almost 3 days and the loss value was 0.0063382 on arctic dataset. Now when I am using the model, the quality of the sound is not good and not clearly audible. Can you please suggest what I can do to improve it? @andabi

Can you please share your Model ??

Amey1109 avatar Mar 04 '20 06:03 Amey1109

Are you asking to share the trained weights or the code ? … On Wed, Mar 4, 2020 at 11:55 AM Amey1109 @.***> wrote: [image: loss] https://user-images.githubusercontent.com/54697554/69516482-998c8400-0f73-11ea-8d64-1d963f7fa5e2.JPG I trained the net2 for almost 3 days and the loss value was 0.0063382 on arctic dataset. Now when I am using the model, the quality of the sound is not good and not clearly audible. Can you please suggest what I can do to improve it? @andabi https://github.com/andabi Can you please share your Model ?? — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#121?email_source=notifications&email_token=ANBJ4UTUKXASEVQDIG7FKNTRFX3M7A5CNFSM4JRD6VX2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOENWS23Y#issuecomment-594357615>, or unsubscribe https://github.com/notifications/unsubscribe-auth/ANBJ4UVUBYWHKMBLHSUFSDLRFX3M7ANCNFSM4JRD6VXQ . -- 1. Hassan khawaja

trained weights

Amey1109 avatar Mar 04 '20 15:03 Amey1109

@hassanShabbir1960 I just want trained weights Can You please Provide it Thanks Alot for replying

Amey1109 avatar Mar 04 '20 15:03 Amey1109

@hassanShabbir1960 Thanks Alot. Have A good Day

Amey1109 avatar Mar 04 '20 15:03 Amey1109

ckpt.zip https://drive.google.com/file/d/12bTbHTcIaUbmedJ7vK-U7vmRgLSiNlP1/view?usp=drive_web

On Wed, Mar 4, 2020 at 8:10 PM Amey1109 [email protected] wrote:

@hassanShabbir1960 https://github.com/hassanShabbir1960 Thanks Alot. Have A good Day

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/andabi/deep-voice-conversion/issues/121?email_source=notifications&email_token=ANBJ4UVVTQLXN7BYWGCU6LLRFZVN3A5CNFSM4JRD6VX2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOENYKZDY#issuecomment-594586767, or unsubscribe https://github.com/notifications/unsubscribe-auth/ANBJ4UVRXKM5CQWXMXMDJNDRFZVN3ANCNFSM4JRD6VXQ .

--

  1. Hassan khawaja

hassanShabbir1960 avatar Mar 04 '20 15:03 hassanShabbir1960

@hassanShabbir196 Thanks Alot You just Saved my day!

Amey1109 avatar Mar 04 '20 15:03 Amey1109

Hello @hassanShabbir1960 As Im new to python Can you please tell me how do run the Convert.py I have the path of the target speech and im getting only 1 sec of converted speech and that too in distorted voice Please Help !

Amey1109 avatar Mar 04 '20 15:03 Amey1109

Hello @hassanShabbir1960 may I have access to your checkpoints too. Will really appreciate it

meharbhatia avatar Apr 03 '20 14:04 meharbhatia