What happens to the sound
Speeding up audio naively raises its pitch — the chipmunk effect — because playing samples faster raises every frequency. Pitch-preserving time stretching analyses the audio and repeats or removes short segments instead, which keeps voices sounding like themselves.
It is not free: beyond about 2x, stretched speech acquires a slight warble. For a lecture at 1.5x it is unnoticeable; at 4x it is obvious.
Slowing down is limited by the frame rate
A 30fps video at quarter speed shows 7.5 unique frames a second, which reads as stuttering rather than slow motion. Real slow motion is filmed at a high frame rate in the first place — 120 or 240fps — so there are frames to spare.
Frame interpolation can invent the missing ones, and it produces artefacts around anything moving quickly. Filming at a higher rate is the only route to clean slow motion.
Frequently asked questions
Will the voices sound strange?
Not with pitch preservation on — it stretches time rather than resampling. Beyond about 2x a slight warble appears.
Why does slow motion look choppy?
There are not enough frames. A 30fps clip at quarter speed has 7.5 unique frames a second. Real slow motion is filmed at 120fps or more.
Does speeding up shrink the file?
Yes, roughly in proportion — half the duration is about half the data, since there are fewer frames to store.
Something wrong with this tool, or an idea for it? Tell us