slow inference

[luyao-cv]

Is your feature request related to a problem? Please describe.

总共推理要经过四个模型，

1. ppg编码， whisper ppg模型，模型比较大。 这块占比比较大，推理速度要40s。
2. hubert编码，推理速度要10s。
3. pitch音高编码，推理速度要10s。
4. svc_infer，推理速度10s。
   1-2-3是串行的，为了省显存

Describe alternatives you've considered

1

Additional context

1

Code of Conduct

• I agree to follow this project's Code of Conduct

Are you willing to resolve this issue by submitting a Pull Request?

• Yes, I have the time, and I know how to start.
• Yes, I have the time, but I don't know how to start. I would need guidance.
• No, I don't have the time, although I believe I could do it if I had the time...
• No, I don't have the time and I wouldn't even know how to start.

[34j]

It is true that pitch inference and ContentVec inference can be parallelized, but in practice this may have little effect because dio is used in realtime inference. Also, since this repository does not use Whisper, I think you are opening this issue in the wrong repo.