How to isolate/figure out whispers in audio clip?

Karl Semich 0xloem at gmail.com
Tue Jun 29 03:08:56 PDT 2021


>
> To do that you'll need a way of generating similar whispers.  This means
> simulating cover noise overlayed on the DeepSpeech training data.
>

Obviously this idea has the error that the training data is not whispers.
So you'll also want to train something to classify data as whispers or not.

>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: text/html
Size: 934 bytes
Desc: not available
URL: <https://lists.cpunks.org/pipermail/cypherpunks/attachments/20210629/5e536ac9/attachment.txt>


More information about the cypherpunks mailing list