“ a much simpler version and will upload soon.”，can this version run in realtime in cpu? about wavernn

[fatchord]

@hdmjdp I'm still working on the simplified model and the results are pretty good so far (more params than your 0.95M though) ... also experimenting with batched inference on GPU.

[bliep]

@hdmjdp "I have changed your model to smaller one(only has 0.95M parameters)."
Would like to have a look, want to share that model?

[hdmjdp]

@fatchord in the picture, your can see several bars, how can i remove it?

0002_wavernn.zip

[hdmjdp]

@bliep it is just for commercial data. and it needs 20Gflops, but I think it can not run in realtime for cpu.

[fatchord]

@hdmjdp I updated the current model in the repo to generate realtime on gpu. Check out notebook 4b.