How to Record EVP: Setup, Recording and Review



To record an EVP session, make a test recording, place the microphone where it will stay still, record continuously, and keep notes about people and background sounds. Back up the original file before reviewing it. Clear audio and a reliable timeline make a recording easier to assess; they do not establish paranormal activity.

EVP, short for electronic voice phenomena, describes recorded sounds that listeners interpret as unexplained voices. This guide focuses on how to record EVP sessions carefully, from setup to review. For the equipment basics, start with what an EVP recorder is and what it actually records.

Affiliate disclosure: As an Amazon Associate, HauntGears earns from qualifying purchases. Retailer links on our product pages may earn us a commission.

EVP recording tips at a glance

  • Test your actual recorder or phone app before visiting a location.
  • Use linear PCM WAV when available, but keep the original format if your recorder only supports MP3.
  • Keep microphones clear of clothing, cases, wind, and vibrating surfaces.
  • Record room tone and note ordinary sound sources.
  • Review the untouched recording before processing or suggesting words to another listener.

1. Choose a recorder you can operate confidently

A phone can be a practical starting point. Check whether its app can record for the session length you need, keep recording with the screen locked, and export the original file. Make a trial recording and play the exported file on another device. Calls, vibration, storage limits, app processing, and battery use can affect the result.

A dedicated recorder can offer physical controls, flexible placement, and headphone monitoring. Features vary by model: a digital recorder does not automatically support WAV, manual gain, an external microphone, or live monitoring. Check the manual before buying.

For example, the Olympus WS-852 records MP3, not WAV. Its manual lists 128 kbps stereo and lower-bitrate mono modes. It can monitor a recording through its earphone jack; changing headphone volume does not change the recording level. Choose it with those limitations in mind.

Olympus WS-852 digital voice recorder with built-in microphones and front controls
The Olympus WS-852 records MP3. A WAV recommendation should not be mistaken for a feature this model offers.

The Zoom H1essential uses 32-bit float WAV and built-in X/Y microphones. Its workflow does not require setting an input gain level, but wind, handling noise, and microphone input overload still need attention. Confirm the retailer’s exact bundle contents and accessories before purchasing.

2. Set recording options for your device

There is no special setting that makes a recorder detect spirits. The useful settings are the ones that preserve a continuous, understandable record of the location. These are starting points, not universal requirements:

Setting Practical choice Why it matters
File format Linear PCM WAV when supported; otherwise the device’s suitable high-quality recording mode WAV can avoid lossy encoding. Converting an MP3 to WAV does not restore discarded information.
Sample rate 44.1 or 48 kHz is a reasonable starting point when selectable A higher number alone does not identify a sound’s source.
Mono or stereo Use the recorder’s appropriate microphone mode Mono can simplify a speech track; stereo can retain differences between channels. Neither is inherently paranormal-sensitive.
Voice activation Continuous recording when you need a complete timeline Activation thresholds can leave gaps or miss quiet beginnings.
Gain and automatic controls Use manual gain if available and suitable; note automatic processing Controls differ by model. Test loud and quiet sounds instead of following a universal switch-off rule.
Low-cut filtering Test it for rumble or wind and log the choice It changes low-frequency content. Reducing noise at its source is preferable when practical.

Make a short test with speech, quiet room sound, and an expected louder sound. Listen back for distortion, rustling, dropouts, or a loose cable. If your recorder has meters, follow its manual’s level guidance. See our EVP recorder settings checklist for a separate setup reference.

3. Prepare the location without compromising safety

Get permission to be there and to record. Explain the session to everyone present, respect privacy, and follow the property’s access rules. Keep exits, lighting, and essential communications available.

Listen before starting. Note nearby conversation, traffic, plumbing, ventilation, radios, and animals. Move away from avoidable noise, silence nonessential alerts, and put loose keys or other rattling items aside. Do not shut down heating, ventilation, refrigeration, alarms, or other essential systems, or seal a room, just to make it quieter.

If you use airplane mode, consider whether you need that phone for communication. It may reduce interruptions, but it does not guarantee a noise-free recording. Confirm the app still works as expected.

Record a period of room tone: the location’s ordinary background sound while the group stays quiet. Note its start and end. A brief sample is useful context, but it cannot document every intermittent sound or prove the room stayed unchanged. Our guide to reducing background noise while recording covers practical noise control.

4. Position the microphone and leave it steady

Use a suitable stand or a stable surface that does not transmit obvious vibration. Keep microphone openings clear. A pocket, bag, loose fabric, or fingers rubbing the case can add muffling and movement noise.

Follow the microphone’s pickup pattern and test the position. There is no single angle that suits every built-in, directional, or clip-on microphone. Keep enough distance from breathing and clothing while maintaining the sound coverage you need. Outside, use a compatible windscreen and find shelter from wind when safe.

Handheld recording is possible, but handling sounds need to be considered. If you move the recorder, note when and why. Headphones can help check a compatible recorder, but keep the volume comfortable and remain aware of your surroundings.

5. Record a session with a usable timeline

  1. State the context. Record the date, time, location, people present, recorder position, and relevant settings. Use a location code if the recording might be shared publicly.
  2. Let the background be heard. Capture room tone before conversation and after a major change in conditions.
  3. Speak one at a time. If you ask questions, leave comfortable pauses so voices and movement do not overlap. There is no established pause length that produces paranormal communication.
  4. Identify known sounds. Briefly say “I moved the chair” or “a car passed outside,” and note the timestamp. Avoid talking over the entire sound if you can identify it afterward.
  5. Keep a written log. Include people entering, devices operating, weather changes, and any recording interruptions. Our investigation log template can help.
  6. Close and check. Stop and save according to the device’s instructions. Confirm the file plays before leaving.

For a clean ambient recording, avoid adding radio chatter, music, or artificial noise. Those sounds make source identification more complicated. If another experiment uses sound playback, label it separately and document exactly what played.

6. Back up and review the original audio

Copy the original file to a second location before editing. Keep its original format, filename, and session notes together. Give working copies clear names, such as room-A_2026-10-22_session-01_review.wav. Avoid repeatedly exporting through lossy formats.

  1. Listen at normal speed first. Use a quiet setting and a comfortable volume. Note timestamps before repeatedly looping a sound.
  2. Keep context. Listen before and after each candidate clip. Compare it with the log, room tone, and any other synchronized recording.
  3. Describe before interpreting. “Two brief syllable-like sounds at 03:14” records less assumption than writing a proposed sentence as fact.
  4. Invite an independent listen. Share an unprocessed clip with context, without first supplying the words you think you hear. Let “unclear” be an acceptable response. See our blind EVP review method.
  5. Keep the conclusion limited. Agreement between listeners does not establish the speaker or cause. An unidentified sound remains unidentified.

A spectrogram displays frequency energy over time. It can help locate a click or compare passages, but it cannot determine whether a sound is paranormal. Display settings can also change how a pattern looks.

Use audio cleanup cautiously

Audacity’s Noise Reduction is designed mainly for relatively steady noise. Its manual explains that variable noise and noise overlapping the desired sound are harder to remove. A noise profile should contain representative noise without speech you want to preserve.

Work on a copy, preview small changes, and compare against the original. Noise reduction can remove useful sound or introduce artifacts that resemble tones or syllables. Keep a record of the processing steps; a cleaned clip is an interpretation aid, not a more authoritative original.

Before you press Record

  • Permission, safe access, and a communication plan are in place.
  • Battery, storage, recording mode, and file export have been tested.
  • The microphone is clear, stable, and protected from avoidable wind.
  • People and background sound sources are logged.
  • Continuous recording and room-tone capture fit the session plan.
  • Original files will be backed up before review or processing.

Frequently asked questions

Can I use my phone to record EVP?

Yes, you can use a phone to record the location’s audio. Test the app, file export, battery use, and interruptions first. A dedicated recorder may make placement and monitoring easier, but neither device establishes paranormal activity.

Is WAV better than MP3 for EVP sessions?

Linear PCM WAV is useful for preserving audio without lossy encoding when your device supports it. MP3-only recorders can still document a session; keep their original files and understand the limitation. A WAV conversion cannot recover information already lost in MP3 encoding.

How long should I pause after a question?

Leave enough time to avoid overlapping speech and movement. A consistent pause can organize the session, but there is no verified number of seconds that makes an unexplained response more likely.

How can I tell whether a sound has an ordinary source?

Check surrounding audio, notes, known noises, microphone handling, and independent recordings where available. Another listener may help identify alternatives. If the source remains uncertain, label it uncertain; lack of an explanation is not proof of a ghost.

Equipment and audio references

Share post


Discover more from Haunt Gears

Subscribe to get the latest posts sent to your email.

Leave a Reply

Your email address will not be published. Required fields are marked *

more insights