Are there significant differences in the air exhaled (both in volume and speed) between languages?
I would doubt it, and I would worry that it relates to individual anatomy more than the language itself. There are a lot of factors at play, and to my knowledge no studies of exit airspeed have been conducted. Linguists frequently study airflow which is volume per unit time, but speed is dependent on a lot of anatomically specific factors.
While many studies look at nasal airflow, these rates are usually for highly controlled environments rather than free speech, and the flow is usually only out of the nose rather than total flow out of the vocal tract. In running speech, the volume of air expelled may differ by context. For example, if you're speaking faster, each sound is articulated for less time so the amount of air exhaled per sound would likely also be reduced. Towards the end of phrases, your lungs are low on air, so if you're forcing it, you could probably get by with less air at the ends of utterances compared with the beginnings.
Determining the exit airspeed for a sound is also a hard question. Airspeed relates to the size of the opening, the pressure difference, and fluid viscosity (moist air has lower viscosity than dry air, for example). Air pushed through a narrow opening will be faster than air through a wider opening, all else being equal. However air pressure is modulated during speech, so air may not be pushed through at the same rate. Different sounds require certain pressure differentials which will affect exit air speed in a way that may overcome lip rounding. For example, the aerodynamic voicing constraint details the pressure difference required for voicing, but even in this case it is a pressure difference, not an absolute pressure, and the pressure a speaker generates with their lungs may be idiosyncratic. We may expect voiced sounds to have faster exit airspeeds than voiceless sounds on average, but any cross-linguistic generalizations will likely be dominated by anatomical differences.
