TODO list: if you have plan to support arm platform hardware accelerate, like arm executorch/xnnpack.
I have one quick setup on RK3588 platform, which includinng 4+4 arm v8 CPU cores. There have one positive result with good RTF on arm CPU.
Want to know if there have plan to optimize performance on arm CPU, based on edge AI.
What do you mean? You can just set device=CPU, or is there a way to make a X100 speed up?
Hi, @yukiarimo,
Thank you for your quick response. I deploy Melo TTS on edge device, and the RTF reach to ~2.0, under Chinese(mix English) condition. That is great. Besides, I find that CPU usage is ~50%, both on arm an x86 cpu. So, I expect to know if there have H/W accelerate solution to use more CPU resource, to improve real-time performance? Thank you!