Compiling deepspeech and tensorflow from source

[JRMeyer]

>>> rishabh.shah0330
[December 21, 2017, 11:14pm]

Hi,

I took the latest source for tensorflow and DeepSpeech. Followed the
instructions to create generate binaries for both but am still seeing
the following ERROR,

> type 'select' is not iterable slash
> ERROR: error loading package 'native_client': Package 'native_client'
> contains errors

I checked that this error is reported in Issue
thread and it has
been closed. However, I am still seeing the issue and not sure how to
move forward.

Please let me know if you need any more info from my side.

Thank you slash
Rishabh

[This is an archived TTS discussion thread from discourse.mozilla.org/t/compiling-deepspeech-and-tensorflow-from-source]

[JRMeyer]

>>> lissyx
[December 21, 2017, 11:34pm]

.

That error is typical of building DeepSpeech against upstream tensorflow
instead of https://github.com/mozilla/tensorflow/, branch master.
Can you make sure you are building against our fork?

[Archived Post]

[JRMeyer]

>>> rishabh.shah0330
[January 2, 2018, 7:10pm]

,

Thank you very much for your prompt reply. This helped me to move ahead
and I am now able to build DeepSpeech locally.

Rishabh

[Archived Post]

[JRMeyer]

>>> lissyx
[January 4, 2018, 12:24pm]

Thanks for the feedback. Do you mind updating your original topic title
with the error to help people fighting the same kind of issue ? Thanks !

[Archived Post]

[JRMeyer]

>>> rishabh.shah
[January 15, 2018, 9:34pm]

,

I am not able to retrieve my username rishabh.shah0330 and hence not
able to edit the topic. Tried few things but not able to get through.

[Archived Post]

[JRMeyer]

>>> guessous.amine
[January 28, 2018, 2:35pm]

How can I install your specific fork on linux having a gpu ? pleaase

[Archived Post]

[JRMeyer]

>>> lissyx
[January 28, 2018, 7:48pm]

Which specific fork are you talking about ? It's already linked in the
README.md how to get and install from TaskCluster.

[Archived Post]

[JRMeyer]

>>> guessous.amine
[January 29, 2018, 10:25am]

Thank you
![:slight_smile:](

[Archived Post]