How to Measure and Test Microphone Frequency Response

Before you start

Frequency response describes how sensitive a microphone is at different frequencies. A flat response keeps similar output across the tested range, while a shaped response emphasizes or reduces specific areas. Measuring this requires an input device, an output device, and a way to compare the signal after it passes through the microphone and any attached interface. The process runs locally on your equipment; no online service is required.

Use a playback speaker that covers enough of the range you need. For a rough voice-band check, a compact monitor may be sufficient; extended high-frequency checks typically need a speaker with a wider specified range. Keep the microphone on a stable stand and maintain the same distance for repeated measurements.

Set the interface to record cleanly without clipping. Reduce audible background noise before the test. In recording software, choose a sample rate that supports the highest frequency you want to inspect: a 44,100 Hz or higher sample rate is a practical starting point. Set the input level so the loudest test signal remains below the clipping point.

Step-by-step instructions

The steps below describe a comparison method: play a known sweep or tone through a speaker and record the result through the microphone. The recorded response reflects both the speaker and room, so avoid treating it as a pure microphone measurement unless those parts are known or controlled.

• Place the microphone and reference point at the same distance from the speaker.
• Set both to the same input path and sample rate where possible.
• Start with a short test tone around 1,000 hertz and set the input level without clipping.
• Record a sine sweep from low to high frequency.
• Keep the playback level unchanged for the complete sweep.
• Stop recording and save the file for analysis.

If you can compare two inputs, route the speaker feed to a reference input device and the microphone to a second input device at the same time. Record both passes or use a direct electrical loopback as a baseline. This makes it easier to separate speaker variation from the microphone result.

Recorded signal level can shift with room reflections. Repeat the measurement without moving the microphone, then compare two runs to see which features stay consistent and which are reflections or background noise.

Check the output

Open the recording in analysis software that can display amplitude against frequency. Confirm the sweep starts and ends at the intended frequencies and that the loudest point did not clip. A clip may appear as a flat or distorted section and can hide the true response.

A usable measurement should show the general shape of the response without unexplained gaps. A sharp narrow peak may be a room resonance, while a broad slope is more likely to relate to the microphone or speaker. If you used a reference input device, compare the microphone curve with the reference curve and look at the difference.

The result is relative, not an absolute calibration, unless you used a known reference microphone or a calibrated input. Treat very low and very high frequency results cautiously because small speakers, room noise, or anti-aliasing limits can reduce reliability at the edges of the tested range.

Common problems and next steps

High background noise creates ragged low-frequency results. Close windows, switch off nearby fans, and place the microphone away from the computer unless the test requires that position. If the recording level is very low, raise the input level without allowing the sweep peak to clip.

A speaker with uneven response can dominate the measurement. Test with a different output device or restrict the result to the range where the speaker is known to be smooth. Changing the distance can also reveal which peaks are room-related because their frequencies shift as the microphone moves.

If the response drops sharply at a specific frequency, check the recording sample rate and playback device. A sampling limit can remove high-frequency content without lying about the microphone. For a consistent comparison, rerun the test at the same sample rate after any recording or playback setting changes.