Podcast Audio Enhancer
This enhancer cleans up spoken-word recordings. It filters out low rumble, reduces steady background noise, evens out speaking levels, and normalizes loudness to the level Apple Podcasts recommends. It exports MP3 or WAV, and the recording never leaves your device.
How to clean up podcast audio
- Choose the voice recording. Clean voices before adding music, because noise reduction also changes music.
- Pick a noise reduction strength. Start with Gentle or Balanced. Use Stronger only for obvious noise.
- Keep rumble filtering and level balancing on, export, and compare with the original on headphones.
What it fixes
It reduces steady noise from fans, air conditioning, and computers, as well as microphone hiss and low rumble from traffic, desks, and handling. It also evens out a speaker who leans away from the mic or laughs loudly.
It can’t fix echo or room reverb, sudden noises like dogs or keyboard clicks, two people on one microphone, music behind a voice, or a recording that clipped. Echo has to be fixed when recording. Use a smaller room with soft furnishings, and keep the microphone 5 to 15 cm from your mouth.
What each setting does
- Rumble filter. Removes sound below 80 Hz. Adult speaking voices start around 85 Hz, so speech is untouched.
- Noise reduction. Lowers the steady noise floor by 6, 12, or 20 dB. Too much makes voices sound watery or robotic.
- Level balancing. Applies gentle 2:1 compression, then normalizes loudness.
Normalization runs in one pass, so results can land slightly off target. If your network has strict delivery specs, check the file with a loudness meter.
Compared with AI enhancers
AI tools like Adobe Podcast Enhance rebuild speech with a trained model. They can rescue a poor recording, but they sometimes change how a voice sounds, and they need an upload. This tool uses standard audio filters. It never adds sounds that weren’t recorded, and it works best on a decent recording with steady background noise.
Frequently asked questions
Can it remove echo?
No. It reduces steady noise and evens out levels, but it can’t remove room reverb.
Should I export mono or stereo?
Mono for a voice recording, which uses the −19 LUFS target. Stereo if the file includes music, which uses −16 LUFS.
Is my recording sent to a server?
No. Every step runs on your device.
