[FEATURE] adopt dance model speed based on music from apps

[maxoptimus]

Describe your idea

maybe use some fft or equalizer lib that can return music intensivity;
trasform it add change dance speed on the tly;

Why is this useful?

can be cool; especial with option to setup default dance file

Confirmation

• I confirm this request includes only one feature.

[pongsathorncha]

Hey @maxoptimus, thanks for opening this feature request! 👋

I am just also a fan of the software, actually had the same idea and worked on an implementation in my fork of Mate-Engine. I also opened PR #566 to propose it back to the original project for review by the original creator.

The implementation is focused on synchronizing the avatar’s dance animation speed with the music by detecting local audio peak / BPM behavior in real time. It then dynamically adjusts animator.speed, so the avatar can speed up or slow down with the song instead of dancing at one fixed speed.

A few notes about how it works:

• It builds on the existing app-isolation system, so the BPM logic reads from the selected/allowed audio app rather than random system sounds.

• The changes are kept inside AvatarAnimatorController.cs to keep the footprint small.

• It uses an energy-flow / rolling average approach instead of strict metronome-style detection, which makes the movement feel smoother and less robotic.

• It includes a short dance-stop grace period to prevent stuttering during sparse intros or brief gaps in the music.

• Very fast tracks automatically fall back into a half-time groove, so the avatar does not look like it is being fast-forwarded.

The PR is here:

PR #566

While the PR is waiting for review, I also made an unofficial test release from my fork for anyone who wants to try it:

👉 https://github.com/pongsathorncha/Mate-Engine/releases/tag/x3.3.0-bpm-v1.2

This should cover the dance-speed adaptation part of your request. The exact final implementation may still depend on how the original project wants to handle it, but the feature is already working in my fork if you want to test it and share feedback.

[maxoptimus]

here https://github.com/SilenZcience/CAudioVisualizer also exists sample how use free fft lib; will be cool to compare BPM and fft results;