add hints for building ctcdecode on unsupported platforms

This commit is contained in:
NanoNabla 2021-01-21 10:52:26 +01:00
parent b9aa725900
commit aec81bc048

View File

@ -327,3 +327,33 @@ much performance gains do you get in your applications, how you had to change
the model to make it work with a delegate, etc.
See :ref:`the support / contact details <support>`
Building CTC Decoder for training on unsupported platforms
----------------------------------------------------------
Alongside linux on x86 and arm, macOS and windows there also exists a lot of
other architectures which are not offically supported by the DeepSpeech team.
Nevertheless, we offer some hits for other architectures provided by users.
If you have problems, ask in our `discourse <https://discourse.mozilla.org/>`.
Feedback on improving this section or usage on other architectures is welcome.
Since, * `SWIG >= 3.0.12 <http://www.swig.org/>`_ does not include our patches please use
https://github.com/lissyx/swig/tree/taskcluster for building SWIG from source.
You can supply your prebuild SWIG using ``SWIG_DIST_URL``
Moreover you may have to change ``PYTHON_PLATFORM_NAME`` corresponding to your platform.
.. code-block::
# PowerPC (ppc64le)
PYTHON_PLATFORM_NAME="--plat-name linux_ppc64le"
Complete build command:
.. code-block::
SWIG_DIST_URL=[...] PYTHON_PLATFORM_NAME=[...] make bindings