[feature]Real-time audio processing

[ChenShayZeak]

Hello, I currently have a requirement for real-time audio processing, such as changing a user's voice to a child's or robot's voice.

I've tried several methods with very poor results. I'm unsure if this is a bug or a functional issue. If possible, please provide some guidance:

1. I tried inheriting WebRTC's ExternalAudioFrameProcessing in the native layer to process the audio through this public interface. I used linear interpolation, but it only works on some phones, and the noise reduction isn't very good.

2. I tried inheriting TrackProcessor and overriding it in RoomOption, but this didn't work.

I would like to know:

a. What solution should I use to handle this requirement? Would using SoundTouch be the best solution?

b. How can I avoid compatibility issues across different devices?

I would be extremely grateful for any guidance from your team. Thank you.

Looking forward to your reply.