Building Deep Speech and Tensorflow for ppc64le

I have an IBM server which has ppc64le architecture which currently has ubuntu 16.04 but I plan on upgrading to ubuntu 18.04. Has any successfully built deep speech and tensorflow for this platform and architecture since there is no release available for this?

We already have an issue about that on Github. Long story short: we don’t have the hardware, so we can’t do it, but you are welcome to try and share. We’ll help you as much as we can.

I know upstream TensorFlow builds and runs on ppc64le, so it should be fine, and deepspeech already runs on amd64, armv7 and arm64, so you could expect some success.

1 Like