Shrikant Joshi’s Post

View profile for Shrikant Joshi, graphic

Independent Audio Creator, Radio Professional, Voiceover Talent, Geek - overall, a very curious guy.

Will AI take away audio-editing jobs? Of course, the answer is yes. But, only eventually. Let me explain. Most of the AI-based audio tools that are being created today are trying too hard to replace existing workers. I personally believe that is the wrong approach to take. See, when the printing press was created, it didn't replace writers. It replaced the effort of writing the same thing over and over again. Photoshop was created, it didn't try to replace the artist. It replaced the artist's tools. And that's the issue with some of these wonderful AI tools in #audio, such as Descript, Adobe Podcast Studio, and such. Or even Adobe Enhance, Resound, Xound, Auphonic etc. "This AI takes over all your podcast editing!" "This AI will edit your podcasts in 30 seconds!" "This AI cleans noise in under a minute!" "This AI writes and composes an entire song - music AND lyrics!" Except the output isn't 100% perfect, so you end up going back to try and patch the bits that the AI missed. And, what would have taken you 30 minutes, now takes 60 minutes, maybe 90. Instead of reducing work, the AI now ends up ADDING to it. Because you now have to take the AI-provided output, deconstruct it, bring it into the editor of your choice, *cough* Reaper *cough* make the necessary edits, ensure it matches with the rest of the audio, master it, and finally re-render. It would have been much simpler if the output from these audio-editors were given to me as a session file for the editor of my choice. Or as a simple VST plugin with several knobs and dials to tweak inputs and weights. But, no, these insist on delivering the final product. Because they want to go straight to the content-creator.  Because they want to reduce the content-creator's reliance on editors. Because they think they are competing with editors. But, you know who's using and paying for those tools the most? THE FRIGGING AUDIO-EDITORS. All these AI tools are focusing on the wrong customer. They are trying to eliminate the VERY people who would be willing to pay for the features they are building. And to all of them I only have ONE thing to say: Pivot. NOW. Look at your customers. Look at who is paying for it. Look at who is using your product. Look at how they are using it. The ONLY question that should matter when designing an AI tool is: "Does it help speed up a task in your customer's workflow?" ...because that is the only question that the customer is asking when evaluating anything for purchase. (NB: First posted on my personal blog here: https://lnkd.in/dr2cHzck)

Certainly, the concern that AI might replace jobs in audio editing is understandable, but it’s important to see AI as a tool that enhances rather than replaces human expertise. As the creator of Xound,, I've prioritized simplicity and ease of use, ensuring that even beginners can achieve professional results effortlessly. This approach makes it especially beneficial in markets with limited access to expensive editing tools, ensuring everyone has the opportunity to create high-quality audio. AI can be viewed as a complement to human skills. Listening to user feedback is crucial; for example the launch on AppSumo provided me invaluable insights, particularly on adapting our product for global markets. Over 4000 new customers asked 135 questions in public, provided feedback and left reviews. https://appsumo.8odi.net/nL3ePX

Will Rice

I help people and organizations get their message out! Communications, Social Media, Voiceovers, Production, Podcasting, Audio Consulting

2w

These tools are amazing. As an editor, AI saves me a ton of time when working with clients with poor audio quality. On the other hand, practically, I lose about one client a month to AI editing tools. Some come back when they realize that the money saved isn't worth the time they have to invest. Others reach out again when they realize that they don't yet sound quite as good. Some people have the time and like being able to do the editing work without the traditional skillset. It will be interesting to watch.

David Hooper

I help people (and companies) build an audience, attract clients, and make money via podcasting and broadcast radio.

2w

I gave Descript (editing) a try, thinking it would help me cut down on bulk, like false starts, tangents, etc. It ended up adding to it, since I still had to listen to the audio of the "finished" content afterward and that's where the traditional editing took place. It's not bad, overall, but you can't edit audio like a written document, because it's not consumed that way. Still, I like that it's made it easy for people to jump into audio with something familiar.

Rohit Kokane

Award-Winning Podcast Host 🎙️ | I help craft personalised podcast strategies and concepts that connect and engage your audience | Finance Executive | Blog Writer

2w

Used Descript and the more we use those audio cleaning features, it reduces the real voice feeling from it. Something that I didn't like.

See more comments

To view or add a comment, sign in

Explore topics