[basve]

And I should add that this was measured using a downsized model (just two blocks of dilated convolutions and a sampling rate of 4khz). Deepmind's paper does not report how many stacks are used to generate the samples, but I assume it's quite a bit more.

[unlikelymordant]

They (deepmind) reported it took 90 minutes of processing to generate 1s of speech via tweet. Hopefully this comes down in the future.

[espadrine]

Do you have a link?

This implementation says: “A Tesla K80 needs around ~4 minutes for generating a second of audio at a sampling rate of 4000hz”, which is significantly faster.

[basve]

90 minutes for 1s of audio was reported by someone from Google on twitter, but the tweet has been deleted. I've clarified in the readme that my measurements are for a much lighter/smaller model than Deepmind's :).