What does happen, I imagine, is that when listening to that mess, we focus really hard on singing along in our heads, and listening for the melody/rhythm. When we can align our internal Freddy to the melody in that recording, it connects in a way that might feel to some like hearing his voice.
I am a huge Queen fan, but I have a phenomenal knack for not being able to remember lyrics or songs. My children upon hearing a song once can sing along better the second time than the 100s of times I’ve listened to most Queen songs. So I believe I am in a unique position to lend my voice here. It is truly in your head- the only parts where I get a true hint of a voice are at the (high) sustained notes at the end of a phrase. It is not clear enough to me during the other parts to assign a lyric to the words.
Perhaps a way to test it for yourself is to jump forward or backward randomly to a new section and see how long it takes before Freddie comes back into your head?
(A piano being used to mechanically reproduce a voice. While reading the words it’s “saying” you can make them out pretty well. That guidance is definitely needed though.)
What is the link in the OP exactly? Is it trying for the same kind of thing or something?
OK. I’ll bite. How is this done? Because when I listen to the Bohemian Rhapsody one (not on Don’t Stop Me Now) I can hear the singing and, I guess I’m either overly-skeptical, a sucker, or both, but, it seems to me there is just a track down underneath of the actual singing–not just the MIDI piano.
Is there info anywhere on what this is all about? How it’s done, etc.?
I didn’t look much into it on how is it done, however the first one I linked on is a MIDI file, MIDI files only contain the musical notes for a computer to play, there are no waveforms recorded so it can’t be that they sneaked the vocals under the key barrage.