Meta releases AI tool that turns text commands into cleaner audio recordings

Sound editing magic is here

Imagine cleaning up a noisy recording with just a sentence. That’s the power of Meta’s new SAM Audio, a free AI tool. It can often isolate unwanted sounds using a simple text prompt, and Meta provides a free demo and code, though processing speed and results depend on file length, model choice, and computing resources.

This technology is changing the game for podcasters, musicians, and everyday video creators. You can now achieve clean audio without any technical skills. It removes the biggest hurdle for creating professional-sounding content.

Type what you want gone

You communicate with this tool using plain English. Just type a phrase like dog barking or background chatter into a prompt box. The AI scans your audio file, finds every match for your description, and separates it.

It neatly extracts that specific sound, leaving everything else intact. The web demo is responsive for short clips, but processing time increases with longer audio files and when running larger model variants locally.

Click a video to isolate sound

The magic gets even cooler when you work with video. Click directly on an object or person in the video frame. If you click a guitar in a concert clip, the AI will isolate the guitar track.

This visual prompting links what you see with what you hear. It’s an incredibly intuitive way to edit, making you feel in direct control of the audio.

Mark a moment in time

Another method is called span prompting. You simply highlight a short piece of the audio waveform where your target sound occurs. The AI learns the acoustic signature from that sample.

It then finds and removes every similar instance throughout your entire recording. This is perfect for eliminating repetitive noises like coughs or door slams.

A gift for music lovers

This is a revolutionary tool for music fans and creators. SAM Audio can often isolate vocals from a mixed recording for practice or personal use, but results vary and users should respect copyright and licensing rules when working with commercial music.

It democratizes access to musical components. Remixing and learning become possible for anyone with a favorite track.

Podcasters, rejoice

Recorded an interview in a noisy coffee shop? Simply tell the AI to remove the espresso machine and background music. Your dialogue will sound crisp and clear, as if recorded in a professional studio.

This rescues audio that you might have thought was unusable. It saves creators countless hours of frustrating, manual editing work.

Fix video audio in seconds

Bad audio often ruins otherwise perfect video clips. Now you can easily rescue those moments. Remove overwhelming wind noise from a hiking vlog or loud airplanes from an outdoor interview.

The tool works directly on the audio track of your video files. It can significantly improve many video and audio tracks, but it is not a guaranteed fix for every possible audio problem.

It’s free and open source

In a surprising and generous move, Meta released this advanced tool for free. Meta offers a free web demo and publishes code and model checkpoints under an open source license so developers and researchers can experiment or integrate the models.

This open approach accelerates innovation and access. High-end audio editing is no longer locked behind expensive paywalls.

Forget complex software

This tool completely bypasses the steep learning curve of traditional audio software. There’s no need to understand waveforms, spectral filters, or complex plugins. You describe the problem, and the AI executes the solution.

It focuses on your creative intent instead of technical skill. This dramatically lowers the barrier to achieving high-quality results.

The future of creative tools

SAM Audio points to a future where we converse with our creative software. You will simply ask for edits in natural language. These AI tools are becoming intuitive partners that handle technical heavy lifting.

This shift allows creators to focus more on their big ideas and stories. Technology is finally catching up to our imagination.

How to try it yourself

You can test SAM Audio right now in your web browser. Visit the Segment Anything Playground on Meta’s AI website. Upload a short audio or video file and start experimenting with different prompts.

The interface is designed to be user-friendly from the start. You might fix an old recording in just a couple of minutes.

Unlock your old recordings

Think of those precious videos with ruined audio. Now you have a key to fix them. Save a grandparent’s story drowned out by room noise, or a child’s performance obscured by crowd chatter.

This technology helps preserve your memories with newfound clarity. It brings old, neglected files back to life.

Want to know more about the company behind this free tool? See why Meta keeps paying billions in fines.

Why this changes everything

This tool transforms audio from a technical chore into a simple fix. It empowers every person to be an editor, regardless of their experience. You can achieve great sound for school projects, family videos, or professional work.

The era of struggling with poor audio is finally fading. Welcome to a clearer, more creative future for everyone.

Curious what Meta’s planning next? Read about why Meta backs away from $600B Trump deal.

What’s the first audio clip you’d rescue with a tool like this? Share your ideas in the comments and give this post a like.

This slideshow was made with AI assistance and human editing.

Don’t forget to follow us for more exclusive content on MSN.

Read More From This Brand:

• Apple loses key AI search executive to Meta
• Meta strikes massive cloud computing deal with Google
• Meta and Arm strike a partnership to expand global AI capabilities