More conversions between audio crates base types

[Roms1383]

Hello, and thanks for the initiative!
To start with, I'm no sound designer or audio connoisseur. I just happen to have released, maintained and improved a game modding audio plugin named audioware. My thoughts might not be worth much, as such, but there's still a suggestion that I'd like to make.

When evaluating different crates over a year ago, I finally settled down on kira because of its simplicity and ease to work with. I'm pretty satisfied of this choice but I noticed, especially lately, that different crates offer different features, for example rodio has take_crossfade_with that kira does not have (yet), while kira has certain features that others do not have (yet).

Lately I wanted to experiment with a crate for digital signal processing which looks fabulous, fundsp, but it's quite some boilerplate for example simply to convert kira::Frame / fundsp::hacker::Frame / etc, or between different buffer (and refs) kinds. And here's my point, exposing more conversion means between different crates types could potentially lead to more experimenting, maybe even mixing different audio engines.

Let me know what are your thoughts on this, thanks for taking the time to read me.

[dvdsk]

One issue is that the current crates work in wildly different ways, that makes it very hard to interface them. For example Rodio currently supports something called "spans". These are periods after which the sample rate and number of channels are free to change. As far as I know that is unique to Rodio. It is also, if I may say so, a pretty worthless features. We've been trying to get rid of them for some time. I have a design that might work now I just need time to try implement it. If we succeed Rodio will be a bit closer to Firewheel and Kira.

[Roms1383]

yes I get your point, came to a similar conclusion even when comparing traits over rodio, kira, oddio.
looking forward to what you're working onto!