Now TextMelLoader, TextMelCollate are running on CPU side, why not offload those tasks to GPU for processing?

NVIDIA / tacotron2

Tacotron 2 - PyTorch implementation with faster-than-realtime inference

BSD 3-Clause "New" or "Revised" License

5.12k stars 1.39k forks source link

Now TextMelLoader, TextMelCollate are running on CPU side, why not offload those tasks to GPU for processing? #624

Open Shan19900305 opened 8 months ago

Shan19900305 commented 8 months ago

I find TextMelLoader, TextMelCollate in running on CPU, and all ops used in those two function are very normal, why not offload those tasks to GPU side.