Kelly Davis
Kelly Davis
@dabinat The original idea as I understood it was to start the engine with several language models and one acoustic model then be able to, at runtime, select which language...
I'm putting this in the 1.0.0 project, as that will contain some language model work, which may include this.
> I missed the fact that this was in 1.0... I don't know if this has to be in 1.0. We can, if it's too much, delay it for 2.0.
We're going to delay until 2.0
@MatthewWaller It doesn't appear as if TenforFlow is supported.
@MatthewWaller I'd guess so. Which seems like a large outlay in time and knowledge.
@MatthewWaller Thanks for the update!
@MatthewWaller Did you try CoreML on the [PR](https://github.com/mozilla/DeepSpeech/pull/1463) or on master? I think some, maybe all, cycles should be removed in the PR.
@reuben Can you give @MatthewWaller a preliminary model for the PR to test CoreML?
@MatthewWaller The branch stores the decoder state in the graph in the variables `previous_state_c` and `previous_state_h`. It's a convenient place to store this state info. As far as I understand,...