The honest guide

Am I funny?
Here’s how to actually know.

Every comic asks it, usually at 1 a.m. after a mic. One number cannot answer it. A recording can give you something more useful: evidence of where this room responded, where the set went quiet, and what deserves another test.

Stop asking friends. Count the room.

Your friends may be too kind and your inner critic may be too cruel. The audience response is valuable evidence, but a recording captures it imperfectly: room noise, crowd size, microphone position, and delivery context all shape what can be measured.

That’s exactly what LaughSignal does: it takes a recording (or a YouTube link) of your set and returns a detected response rate, an evidence-labeled response timeline, and three strong moments to verify against the tape.

What is laughs per minute (LPM)?

LaughSignal’s current LPM is the number of audio- or laugh-marker-supported response candidates divided by analyzed minutes. Forty supported candidates across a ten-minute analyzed window is 4.0 LPM. Speech pauses remain timecoded for review but do not change LPM without independent response evidence.

LPM becomes more useful when you compare similar recording conditions. Put two versions of the same material beside each other, inspect the timecodes, and use the number as a lead—not as a final ruling on which room or set was “better.”

Why LPM is one useful scoreboard

• It is inspectable. Every number points back to timecoded moments and an evidence label you can check.
• It is repeatable. The same formula lets you compare versions while keeping room and recording differences in view.
• It can reveal pacing changes. Tightening a setup or cutting a quiet stretch may change response density; the tape tells you why.
• It ranks moments within a set. Set-relative duration and intensity help you find likely review and clip candidates without claiming a universal joke score.

What’s a good LPM?

There is no universal pass/fail number that survives every room, crowd, microphone, and style. The most useful benchmark is your own material across similarly recorded sets. If the same five minutes rises across several nights, inspect the response map and tape to learn what changed before you credit the writing, delivery, or room.

Why clean audio is non-negotiable

Here’s the catch that trips people up: the measurement is only as honest as the audio. LaughSignal reads crowd response from the sound of the room, so how you capture that sound decides how accurate your number is.

• Get the recorder near the room, not just the mic. A soundboard feed captures your voice but often misses the audience. A phone or recorder placed to catch the crowd hears the laughs.
• Don’t clip the levels. Audio that’s peaking and distorted smears speech and laughter together. Aim for clean, un-maxed levels.
• Mind the noise floor. A blender at the back bar, an AC unit, or table chatter can mask real laughter or masquerade as it. The quieter the room noise, the sharper the read.
• Longer than three minutes, shorter than ninety. Enough set to measure, not so much that nothing is comparable.

When a recording is too rough to score honestly, LaughSignal says so instead of inventing a number — and a rejected upload never costs you your free monthly set. Clean audio in, honest signal out.

How to actually get better (with data)

Measuring one set tells you how tonight went. The real gains come from tracking the same material over time: watch a joke’s laugh grow or shrink night to night, see whether your setups are too long, and catch the dead stretches costing you momentum. That’s the difference between doing reps and doing deliberate practice.

LaughSignal sorts detected moments into set-relative tiers — Bankers, Hitters, Workhorses, and Reworks — to prioritize tape review, not to grade the writing. With the Analyst add-on, you can compare similar setup excerpts across nights, then verify whether the same bit is actually growing.

Questions comics ask