Replies: 1 comment
-
Yes, the output is identical. See this discussion over on the NeuralAudio repository: It is primarily about performance, but the test tool also computes error between the implementations. For WaveNet models, you can see than NAM vs Internal (the NeuralAudio implementation) error is zero. For LSTM models, there is a very small amount of difference do to slightly different initialization conditions. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Maybe it's a stupid question but are NeuralAudio and NAMCore identical in null test performance? We know from Leo Gibson that NAM is the best solution at this moment, I'm wondering if with Neural Audio this is the same.
Beta Was this translation helpful? Give feedback.
All reactions