When I teach forced approaches, I teach them to a landing (not just an overshoot), and I remind the candidate pilot that the speed for "best glide" presented in the flight manual will be the speed which takes you the greatest distance for altitude lost, it might not be the ideal speed for the final segment of your gliding approach, that may be comfortable a little faster. Your first clue to this will be the relationship between "best glide speed" and Vy for your aircraft. The faster of these speeds will be the better speed for the final stage of your approach. You can always slip off a few too many knots if you're too fast, stretching a glide is much less certain. And remember that it takes a little bit of your stored energy (as speed) to flare, so don't arrive to the top of the flare gliding close to your stall speed, as you will just thump in when you pull to flare.
The only exception I can think of it if you were clearly in an environment that could kill your engine - heavy icing, or ash cloud if you're a big plane - where descending out of it dramatically improves your restart chances.
What type of aircraft, what is standard cruising altitude (are you dealing with pressurization & potentially oxygen problems?), does the engine have a starter of some kind or will you need ram air of a certain speed to restart, do you need max glide or do you have excess altitude to a suitable field?
Something to add, it takes a lot to stop most small piston propellers (slow airspeed to stop it). They tend to windmill pretty good if it isn't feathered. I would think it's pretty uncommon to have to dive to spool it up again. Something to assess if it was a sudden stoppage vs a slow spool down.
Another reason why a gliding license is a great asset. Do some reading on polar curves. Sometimes best glide isn't the best. If you're trying to punch through a headwind, increasing airspeed to a certain extent will be more beneficial if you are trying to cover ground. Much like flying flying min sink with a tailwind will buy you some time and distance.
I also recall early on in my training that roads were being taught as not good options. Every. Single. Engine failure. I have put it on a road. I have operated spray planes off roads and I'm very conscious of hazards at low level (wires among many other things), so I'm confident in my ability to avoid hazards. Do what you feel is safest, but I still recommend roads over anything other than an airfield for anyone that asks for my opinion. Obviously use your head...the 401/427 in rush hour over Toronto would be the last place I put one down. Don't choose a portion of road in a subdivision or near farm houses.
I will also note, that not one engine failure of mine in small singles went textbook. None had radio calls, some I didn't have time to cause check. One I didn't even feather. Most of mine have been low level. I think with these numerous personal real life examples; finding a suitable landing area is your number one concern. If you can do that, a successful outcome is increased substantially. After all....it's just a shitty glider at that point, and hundreds of KIDS learn to fly gliders every year in Canada.