The best way to compare sound quality between any two devices is to setup a level-matched (preferably blind) test. This is unfortunately not trivial to do perfectly, but it is absolutely worth learning how to do it!
Our host
@amirm did a nice video on this a while ago which is IMHO worth watching: "
Audio Blind Testing - You Are Doing It Wrong!".
First thing to do would be to try and find a way to switch between the two devices quickly. Auditory memory is short (few seconds) and unreliable, so we really want to keep switching times as fast as possible. Ideally you would play the same track via both devices and use some kind of switching device that lets you instantly switch to one or the other.
Next thing to do would be to level-match the outputs of the SMSL and WiiM. Note that level differences as small as just 0,1dB have been known to skew listening tests results, so you really want to get the playback levels to be as close to identical as possible.
It is important to note that electrical level matching is more precise that acoustical level matching, and as I said above, we need the levels to be very close for a meaningful test. That means that measuring sound levels with a phone SPL app likely won't be precise enough (but it is probably better than nothing).
You can do electrical level matching by playing an e.g. 200Hz sine tone through either device, and measuring its output with a digital multimeter. Then tweak the level of either device until they are as close as possible.
As an alternative to a digital multimeter you can also use your PC soundcard input with the free
Room EQ Wizard (REW)
RTA analyzer tool. As long as you match levels of both devices to <0,1dB you are good! We can use a PC audio input for this purpose because we don't need a calibrated device to match levels; we only need the relative levels to be as close to identical as possible. REW also has the
signal generator tool which you can use to generate the 200Hz test tone you will use to match levels.
If you didn't use similar tools before, it might take time to learn how to use REW correctly, though - audio measurements are not trivial to do correctly and generally come with a steep learning curve!
Once you have a way to instantly switch devices, and you got the playback levels close enough, you can try to make the listening test more reliable by doing it blind - which means that someone else does the switching so you don't know which device is playing at any time. You would need to repeat the test at least 10 times and be able to correctly guess which device is playing at least 9 times to get some statistical validity.
Doing a test in this way eliminates most common sources of bias which means you are comparing just how the two devices actually sound. That will help you determine if there is any meaningful difference in perceived sound quality between the two methods.
Good luck and have fun!