In a way it's a friendly reminder that for some workloads, when latency matters, the GPU might be wrong for the job.

This keeps being a very interesting TTS model. Looks like it's getting simpler to deploy too.

Sounds like a very interesting model (pun intended). It's really nice to pack that much performance in a smaller neural network.

Impressive new models for text to speech.

Now this is a very interesting use of generator models. I find this more exciting than the glorified chatbots.

The newer text to speech models are really good now.

Looks like an interesting engine for offline intent recognition.