Fit a voiceover script to an exact runtime

Revise this voiceover script to fit a target runtime.

Script: [PASTE SCRIPT]
Target runtime: [DURATION, e.g. 90 SECONDS]
Acceptable tolerance: [e.g. PLUS OR MINUS 3 SECONDS]
Reading pace: [WORDS PER MINUTE — USE 150 IF UNSURE; 130 FOR DELIBERATE DELIVERY, 170 FOR ENERGETIC]
Lines that must survive verbatim: [PASTE ANY, OR "NONE"]

Work in this order and show your work:
1. Count the current words and state the estimated current runtime at the given pace.
2. State the word budget for the target runtime and the surplus or deficit.
3. If cutting, remove in this priority: repeated ideas; hedges and qualifiers; scene-setting that delays the point; adjectives that do not change meaning; whole sentences whose information appears elsewhere. Cut whole thoughts cleanly rather than compressing everything — a script ten percent over needs one paragraph gone, not every sentence bruised.
4. If expanding, add a concrete example or one clarifying restatement of the hardest idea. Never add filler phrases.
5. Preserve the opening hook and the final line's function (call to action, resolution, or handoff) unless I marked them cuttable.

Return: the revised script; a timing table with each paragraph's word count and cumulative timestamp; a change log listing every cut or addition with a one-line reason; and the final word count with estimated runtime. If the target is impossible without losing a must-keep line, say so and give the closest achievable runtime instead of silently breaking the constraint.