Back in Austin, out of the car and in front of a proper set with a key light. This checks the Day 6 review’s five asks against the raw take, then sets five for Day 8. Every number comes from the raw take, the original camera audio and the word-level transcript.
Day 6 asks that moved: three fixed, two better
of the frame near-black, down from 34% in the car. The key light did that
lost to 12 stalls of 5 seconds or more, mostly when moving to the next point
speech level, 14 dB under target. The room is quiet, but the mic is too far from you
What the Day 6 review asked for, and what Day 7’s raw take shows. Below them are the long-running items from Day 1.
In order of payoff. The set is fixed, so these are about the mic and how you move between points. The content was strong. The gaps between points are where the time went.
Speech averaged −34.4 dBFS, 14 dB under the −20 target. Your loudest syllables already reach −7.3, so gain alone can safely add only about 6 dB. A lav, or the mic just above the frame, makes up the rest and keeps this quiet room quiet.
12 pauses of 5 seconds or more added up to 2 minutes. They came between points: “What else? What else? What else?” at 7:30, and “moving on to the receptionist” said twice at 10:51. Write one line per point, in order, including the inbox bot you remembered at 15:30.
You said it 5 times, and both swearing restarts (4:00, 7:52) came straight after it. Stop at the full stop. If you slip, pause and say the sentence again without commenting; the edit keeps the clean version.
62 “um”s and “uh”s, 3.3 a minute, up from 2.2. In 15 cases the “um” came right before a pause of over a second, so it signals the stall coming. A silent breath is easier to cut.
At 15:30, “I forgot to mention” led to a 13-second pause and a note to the editor, then the iMessage line came out twice and “more context” three times. Put it in the outline and you’ll say it in the right spot. If it does come late, one line (“this goes after the ads”) and then the point, once.
The biggest visual jump of the series. Your face is almost a stop brighter than in the car, the background sits under you, and the dark share of the frame dropped from a third to about a twentieth.

Median brightness inside each box, 0–255. Near-black share: 5.6% across all 16 sampled frames, down from 34% on Day 6.
New set, same habits: lens at eye height, you centred in a mid-chest frame, steady across all 16 sampled frames.

One frame every 72 seconds, MVI_3184 to MVI_3187, timestamps from the start of the take. Ungraded.
The quietest room of the series (tied with Day 2) and no clipping. The one problem is distance: the mic picks you up 13 dB quieter than it did in the car.
The same measurements on every raw take, Days 1 to 7. Fillers and pauses are counted per minute of talking, first word to last word, so long and short takes compare fairly. Hover or tap a point to see its value.
| Measure | Day 1 | Day 2 | Day 3 | Day 4 | Day 5 | Day 6 | Day 7 |
|---|
Raw takes: 14.9, 9.4, 8.0, 19.0, 6.2, 13.7 and 18.9 minutes of talking. Pace: 162–184 words a minute on Days 1–6, 168 on Day 7. Day 7 also had 0.4 “you know”, 0.8 “kind of” and 2.3 sentences starting with “So” per minute.
Your clearest structure yet: three businesses and tips, each told as getting customers, then the work, then the edit. The content held up. The time went into the gaps between points.
Calmer than in the car, and it suits a breakdown video. The stalls show on your face more than they come through in your voice.
In 14 of 16 sampled frames. One of the two exceptions (5:25) came just as the 14-second stall began.
When you list the three businesses at 0:36, hold up one, two, three fingers. It’s the best moment in the video for it.
“I don’t even need my laptop” and the bot calling you are the fun lines. Let them show on your face.
168 words a minute, your slowest since Day 2. 159 in weddings, 183 in the receptionist part, the topic you know cold.
In the long pauses you murmur and look away. Look down at the outline instead, then come back to the lens and start the line.
The promise 18 seconds after the room tone, and a tidy “see you in the next one. Peace.” Both ends are clean for the first time.
The Austin home studio, from today’s set. Five minutes, every time. Your ticks are saved in this browser.