The hidden background noise that can catch criminals

Tom Scott3 HN recommenders · 2022–2025
5.8M views · explainer · evergreen · Uploaded 2021-12-20

Tom Scott explains Electrical Network Frequency analysis, where hum from the power grid in a recording is matched against grid logs to date and place it. He talks to a researcher and sets them a challenge.

The go-to short, accessible explainer for mains hum forensics, linked whenever recording authenticity or timestamps come up.

Watch on YouTube

8 comments from Hacker News

@spuz2022 · in “Launch HN: Reality Defender (YC W22) – Deepfake Detection Platform”

There are companies that already provide this kind of service so it might be interesting for you to combine efforts: [VIDEO]

@AnotherGoodName2022 · in “Extracting Audio from Visual Information”

If you haven't already seen tom Scott's video on background noise giving the location of a recording due to power line frequencies being subtly different I recommend it. [VIDEO]

RecommendsView on HN
@subpar2023 · in “How to date a recording using background electrical noise”

If this is interesting to you or your kids, here's a less detailed but pretty accessible video [1] explaining how mains hum forensics work, complete with British accent. [1] [VIDEO]

RecommendsView on HN
@Groxx2023 · in “How to date a recording using background electrical noise”

A video on the same subject: [VIDEO] I really enjoy Tom Scott. Everything he puts out is wonderfully concise.

RecommendsView on HN
@LimeLimestone2023 · in “Wearable device that records your voice for legal defense”

According to a Tom Scott's video[1], you can use background noise emitted by electricity to prove that a recording has been taken at a certain time and place. [1]: [VIDEO]

@raphaelj2023 · in “Ask HN: Tell us about your project that's not done yet but you want feedback on”

I'm writing a open-source library ( [LINK] ) and a webapp ( [LINK] ) that detects the Utility Grid noise emitted by electrical devices (A/C, motors, appliances ...) from Video/Audio recording. It then tries to date the recording by matching it to historical data. The signal processing part is getting decent. I'm still trying to collect more reference signals in RU, UK, China and Japan. Tom Scott made a video about the idea: [VIDEO]

@sschueller2024 · in “Ask HN: Are there strategies for content creators to prove their location?”

Probably not practical but there was an interesting video by Tom Scott where investigators used the electrical hum background noise to determine when something was recorded. Maybe there is a way to audio record something to prove location. [VIDEO]

@thewanderer19832025 · in “Hiding secret codes in light protects against fake videos”

Reminds me of the 'mains hum' technique that was used to identify videos. Can also be done with light. [VIDEO] [LINK]