EVP Recorders Explained: How Electronic Voice Phenomenon Recording Works

An EVP recorder isn't a special piece of ghost-hunting technology. It's a plain digital voice recorder, and the interesting part of an EVP session almost never happens while you're recording, it happens afterward, when you sit down and actually listen back closely. Here's what the recorder itself is doing, why the review step is where the real work is, and how to correlate what you hear with everything else that was happening in the room.

1.0What "EVP" Actually Means

Electronic voice phenomenon (EVP) refers to a voice or voice-like sound that shows up on a recording but wasn't consciously heard by anyone in the room at the time. The recorder itself has no way to detect or flag this, it just captures whatever audio reaches its microphone, at whatever noise floor that microphone has. The phenomenon, if there is one, only becomes apparent on playback.

That means the recorder's job during a session is narrow: stay running, stay quiet, and capture as clean a noise floor as possible. Any handheld digital recorder, a phone's voice memo app, or a lav mic into a laptop all do this equally well. There is no hardware feature that makes a recorder better at "catching" EVP, only a cleaner preamp and lower self-noise, which matters for hearing something faint later.

2.0Where a Plain Recorder Falls Short

A standalone recorder gives you one long audio file and nothing else. To find a moment worth reviewing, you have to scrub through the whole session by ear, which is slow and easy to miss things in, especially anything quiet or layered under room noise. And once you do find something, a plain recorder gives you no way to check it against what the room actually looked like at that exact second, or whether anything else in the space registered a reading at the same time.

What it can't do: isolate a moment visually, show frequency content instead of just amplitude, or tell you whether a camera, a thermal sensor, or a field sensor logged anything else at that same timestamp.

3.0Why a Spectrogram Changes the Review

A spectrogram plots frequency content over time as an image instead of just a waveform. Speech-like sound has a distinctive, varying shape across frequency bands, while common artifacts like electrical hum, HVAC rumble, or a distant fan tend to sit as flat, regular bands. Looking at a session as a spectrogram makes it much faster to spot the handful of seconds actually worth a close listen, instead of scrubbing an entire recording by ear.

4.0Reviewing and Correlating in SPECTER

The SPECTER EVP Audio Lab is built around this exact review step. It lays a session out as a waveform and a spectrogram side by side, with a live spectrum readout and an EQ so you can pull a faint sound forward. Drag across the waveform to mark an in point and an out point, and playback loops that region so you can listen closely without re-scrubbing the whole file.

• Everything is timestamped together. A running timecode (HH:MM:SS:FF) overlays the playback, synced to the same global project clock as your camera, thermal, and sensor evidence, so a moment on the audio maps directly to what every other instrument logged at that same second.
• Per-event camera cycling. When an audio moment lines up with a detection event, you can step through the camera views that were active at that timestamp without leaving the lab.
• Export a clean clip. Once you've isolated a region, export it as its own share pack instead of handing someone the whole raw session.

5.0Recorder vs a Correlated Review Workflow

6.0Getting Started

You don't need new hardware to start. Record a session with whatever microphone you already have, load it into SPECTER, and use the spectrogram to find the moments worth a closer listen.

7.0Frequently Asked Questions