All products

Reduce background noise in video with AI

Clean up speech in an audio or video file. For supported videos the agent can return the original picture with the cleaned voice track in place.

Reduce background noise in video with AI when a voice is competing with a fan, passing traffic or the echo of an empty room. The adobe-cleanup-audio skill sends the recording to Adobe’s speech tool and takes the voice track from its results. An audio source gives you audio back.

For a short filmed demonstration, the agent can put that track under the original picture. Listen to quiet sentences as well as loud ones before choosing the take. This is a speech-only result, so room ambience and background music will be absent.

What it does

  • Prepare a noisy interview for a second listen.
  • Reduce room echo on a recorded explanation.
  • Take a separate speech track into an editing session.
  • Replace the sound on a supported video while keeping its picture.
  • Keep the cleaned audio when a longer video exceeds the rendering limit.

Install and use

  1. Install the skill in your agent and connect Adobe’s creativity tools. The source file is processed by Adobe.
  2. Supply a recording you are authorised to use, through an available picker, upload or Creative Cloud asset.
  3. Ask, “Clean the dialogue in this short product demonstration and return the video with the replacement sound.”
  4. Listen for each speaker’s quieter words before choosing the result.

FAQ

Source

Made by Adobe. Original skill: adobe-cleanup-audio, published on GitHub under the Apache-2.0 licence.

View on GitHub