Release of v3.1.0 - Normed Audio Features
This release of Audectra brings some much needed changes to simplify the process of setting up or combining bindings with audio features for your effects.
System Volume Invariance
With this release, Audectra is now fully invariant to changes in the systems master-volume. This time, not just estimations or stuff - but the real thing. This change allows Audectra to be more dependent on your audio playback rather than how loud you are listening to it. Try it out!
Normed Audio Features
With the changes above, I've been finally able to normalize all audio features into a range [0, 1].
Redesigned Expression Fields
The expression fields, which are available for value and trigger bindings, have been redesigned. Here is a list of currently supported fields.
- Spectral.Bins, (each including MaxEnergy, MeanEnergy & Rms fields)
- Confidence (Caution: now ranging [0, 1])
Caution: Because of all the changes above, loading your projects from previous versions auf Audectra will
- strip all state transitions from the project, and
- deactivate all active bindings on your effects.
- Fixed several bugs in TPM2.Net protocol implementation regarding frame index, payload size and frame splitting.
- Massively improved Audectras invariance to system volume changes.
- Small performance improvements in the core loop.
- Changed font in expression editor.
- Normed audio features.
- Redesigned expression fields.
I am making good progress so far with the new documentation. If all goes well, it will be available soon!
After finishing the documentation task above, I will be working on a more refined effect-handling in Audectra, allowing each effect to:
- set its own requirements (like minimum or maximum project dimensions), which will be taken into account when listing all available effects for a project.
- be located in a subfolder path beneath /Extensions/Effects/.