I have used a number of text to speech apps and sites. This is definitely my favorite.

Ultimately what I am most impressed with is the editor.

Compared to other software, the ability to add new speakers to a sound clip is simple (for the few that provide this ability). Essentially, you can write dialogue and interactions between 2 or more voices by adding a line and accessing a simple dropdown menu.

It isn't necessary to learn SSML tags or other markups in order to manipulate pitch/emphasis/rate/etc. Everything is accomplished by adjusting sliders in a drop down after selecting the specific text to manipulate.

Adjustments for pitch/emphasis/rate/etc. can also be done on a block basis instead of selected text, so separate lines can be manipulated easily.

Unmixr also has one of the better "preview" setups. You can listen to previews of small pieces of selected text without having to listen to the entire script. In other words, if you are manipulating pitch/emphasis/rate/etc then you can listen to just that snippet.

If I have one complaint, it is that there is no "undo" function in the editor. If I accidentally delete some text there's no way to get it back, it must be manually typed back in.