Travis Morton

Sure, I'm on commit `67eacb995aef465d1e2ed810f25e0d7d3899e9b6` and this is the loss curve This is one of the generated samples: [gen_LJ001-0028.wav.zip](https://github.com/bfs18/nsynth_wavenet/files/2192844/gen_LJ001-0028.wav.zip) For reference, this is the original that was used for the...

Ah, thank you. My batch size is 2 so I'll continue training. I'll go ahead and close this issue and report back when I have good samples for posterity.

Good catch @bfs18, thank you! I'll try training with `use_weight_norm` as false. Unfortunately my training is a bit slow with a single GPU. Do you have a pretrained teacher model...

Actually, I just saw the updated readme... I'll check that out now.

I was able to confirm that the pretrained model produces a recognizable voice, all that was needed was changing the path in the checkpoint file to a relative one. Great...

Very cool! To my understanding, we should be able to feed these vectors directly into the training and generation, after a sort of preprocessing step where they are generated based...

Thanks @bajibabu! I've never used HTS or merlin, could you walk through the steps you used to create these in more detail?

Definitely on the roadmap! Between rustler precompiled and precompiled libonnxruntime(s), I think there's a lot we can do to make deployment quicker and simpler. In the meantime, we're open to...

I've attached the trace, I don't see anything obvious at first glance: [trace.txt](https://github.com/ninenines/gun/files/12716467/trace.txt)

Thanks! That fixed it for me. For anyone coming here with the same issue, this was my change: ```elixir streamref = :gun.ws_upgrade(pid, '/ws', [{, }]) ```