Speech cleanup studio

Filter hum, rumble and hiss from a local voice recording.

On your device · No account

Privacy details

Speech cleanup studio processes your input in this browser. Inputs and results are excluded from analytics. Recent tools save tool names, visit counts and last-visit times. Saved tools store only their names. Draft saving is optional and stays on this device; delete saved text with the draft controls. Loading text from a URL is optional and contacts that server; its URL and your IP address are visible to the server. An optional “Use in” action keeps a handoff in this tab. The next tool removes it on arrival and ignores it if more than a minute has passed.

Copy link includes the inputs you choose to share in the address. Anyone with that link can reopen them. Files and generated results are excluded. Password and key tools share settings only.

Choose a voice recording

Hum filter

Rumble and hiss filters

High-pass cutoff (Hz; 0 turns it off)

Low-pass cutoff (Hz; 0 turns it off)

Filters affect speech as well as noise. Listen to the processed playback before saving.

Mono or stereo, up to 30 MB and 12 million decoded samples. WAV output.

Processed audio

Preparing example…

Reset reloads the tool with its starting values and releases open files and device controls. Saved drafts stay on this device until you delete them.

Links include your inputs and settings. Anyone with the link can see them, so leave out private text. Files and generated results are left out; random draws run again.

Keyboard shortcuts

Updated 2026-09-20

Filter the frequencies you do not need

A high-pass filter reduces frequencies below its cutoff; a low-pass filter reduces those above it. The optional narrow notch targets 50 or 60 Hz hum. These are ordinary frequency filters, not AI voice isolation. Set a cutoff to 0 to turn that filter off. Listen to the result before saving.

A worked example

With a 60 Hz notch selected, a steady 60 Hz test tone is attenuated after the filter settles. Nearby speech frequencies may change too.

Next: Podcast loudness LUFS studio

Before you use the result

Browser-decodable mono/stereo clips, up to 30 MB and 12 million samples. Output is 48 kHz, 16-bit PCM WAV. Filters cannot separate overlapping voices, remove every background sound or recover clipped speech. The result may sound duller or thinner; reduce the filtering if needed.

Next: Audio compressor

Cite this page

Citation style

ToolOctopus. “Speech cleanup studio.” Updated 2026-09-20. https://tooloctopus.com/speech-cleanup-studio.

Add an access date if your instructions require one.

Citation formatting uses citeproc-js by Frank Bennett and Citation Style Language styles. Licence and source code.