Quinn McNamara

Looks like whats in the public [tensorflow](https://github.com/tensorflow/tensorflow/tree/master/tensorflow/contrib/slim) is lagging a little behind (maybe by design) what is in the models/slim [repo](https://github.com/tensorflow/models/tree/master/research/slim). Seems like the only way to get full functionality...

Potentially useful: https://github.com/tomrunia/TF_FeatureExtraction

Sounds good. Yea I did a split from `speech_recognition` when we wanted to request timestamps from the API. We could probably remove the dependency and write the audio encoding code...

Yep! I think your suggestion is all that's needed! You'll need version >= 2.11.0 of the python sdk too.

@AlienKevin Thank you so much for this thorough exploration and write up! We will start looking into your suggestions. I know that C++17 is top of mind for us and...

Hi Thibault! We still need to work with the podcast creators to get permission to share the raw audios, but this will be the right place to look for them!

Sorry for not getting to this earlier! Added in #45

I can attest to the quality of of Rev.ai speaker diarization, though at the moment it only comes as a package with transcription jobs. 😄 For free/open source, I've also...

Hi @sulutian ! Are you in a directory that is a clone of `reverb`, this installation currently relies on relative paths? It looks you are in a folder named `nnni`....