Audio Editing

Cleaning a recording no longer starts with drawing a noise profile by hand. This category covers repair and finishing: broadband noise and hum removal, reverb reduction on rooms never meant for recording, de-essing, click and plosive repair, automatic levelling so a quiet guest matches a loud one, and loudness normalization to the target a platform expects. Source separation pulls a mix apart into stems; speech enhancement rebuilds a phone-quality take toward something closer to studio capture.

Transcript-based editing belongs here too, where deleting a sentence of text removes the matching audio. Video editors rescuing location sound, podcast producers finishing episodes, archivists restoring tape and lecture-capture teams processing batches rely on audio editing AI tools. Differences worth weighing: how audible the processing artifacts are, single-button versus parametric control, batch capability, plugin availability inside an existing session, and whether processing happens locally or by upload.

Try candidates on your worst recording, not your best. Aggressive denoising strips room tone and leaves a swirling quality behind voices, heavy enhancement makes speech sound synthetic, and separated stems bleed. Confirm that uploading confidential interviews fits your own policies and that originals are preserved non-destructively. Pricing is commonly metered in minutes processed, sold per seat, or charged once as a license for installed plugins.