MikeNet / Soundcheck
Your mic is too quiet. Start here.
Keep a recording before changing anything. You need something to compare against.
1. Check which microphone is recording.
A headset can be connected while the browser still uses the laptop microphone. Start one soundcheck, then inspect the reported input device. If the wrong input is selected, choose the intended microphone and record again. Device names may only become available after permission is granted.
Speak normally near the intended mic. Do not tap or blow into it to test the signal; that tells you little about your speaking voice and can make playback unpleasantly loud.
2. Change distance before changing gain.
Record take A from your usual position. Move somewhat closer without touching the microphone and record B with the same sentence and speaking volume. Listen for clearer words, less room echo and less background sound relative to your voice. There is no single correct distance for every microphone.
If closer speech adds breath pops or a boomy tone, move slightly to the side of the air from your mouth or use the microphone’s intended pop filter. Check its instructions for which face or end should point toward you. Keep the input gain unchanged during this comparison.
A: your current desk position. B: the same setup with the microphone closer. Write “closer to mic” in the change note. Keep your listening volume the same for both clips.
3. Adjust input gain in a small step.
If placement is suitable and the recording is still quiet, increase the input level in the device or operating system. Repeat the test after one adjustment. Listen to louder words as well as quiet ones. Increasing gain raises the captured signal and can also raise captured noise; it does not remove room sound.
On Windows, open Settings → System → Sound, select the input, and use its test and volume controls. On macOS, open System Settings → Sound → Input, choose the device and adjust input volume where available. Some USB interfaces control gain on the hardware instead.
4. Read the meters as evidence, not a grade.
A speech RMS change from −36 to −24 dBFS is a 12 dB rise in average digital signal level. That example does not establish an ideal target. If room tone rises by the same amount, the voice-to-background separation has not improved. If peaks reach digital full scale, reduce gain and compare the sound again.
A near-full-scale sample count flags samples close to the digital limit. Distortion can happen earlier in a microphone or interface, and processing can hide it from that count. Your ears still matter.
5. Confirm it in the app you use.
If MikeNet sounds fine but a meeting or game app is quiet, compare the selected microphone and that app’s input level, automatic gain, voice activation and noise controls. A browser recording cannot prove the other application is using the same path.
Manufacturer guidance on direct and ambient sound: Shure, Microphone Techniques for Recording. System controls: Microsoft microphone setup and Apple sound input settings. The A/B exercise and numerical example are MikeNet’s own.