How Villemure Uses LALAL.AI to Keep Clients Happy & Save Hours on Audio Post-Production

Advances in AI technology are opening new possibilities for mastering professionals to streamline workflows, tackle complex audio challenges, and even maintain creative control. One such tool is LALAL.AI, an AI-powered stem separation and audio enhancement service designed to save time and improve the quality of audio projects.

For this case study, we spoke with Jean-Philippe Villemure, a seasoned audio post-production specialist and mastering engineer who has spent over 20 years helping artists and creators bring their music and audio projects to life.

Villemure shared insights into his work with indie and local musicians, as well as marketing agencies producing video and podcast content and how LALAL.AI fits into his mastering and post-production workflow to save time, solve tough audio problems, and deliver impressive results for his clients.

Villemure: A Solo Studio with a Focus on Mastering & Post-Production

Villemure has been working as a self-employed audio post-production specialist for 20 years. His main focus is music mastering, though he also handles audio post-production for video content and podcasts created by marketing agencies.

Read also: LALAL.AI: AI Solution for Ad Agencies to Craft Unique Video & Audio Content While Keeping Production Costs Low

His work covers mastering albums, EPs, singles, and podcasts, as well as audio for video projects like social media clips and mini-documentaries. Villemure doesn’t specialize in any single genre, often working with the indie DIY scene across rock, pop, folk, hip-hop, country, jazz, and metal.

"Today, I’m working on an album for children; last week, I was on a musical recorded live show.

"First, I talk with the client and go through all the files and any documentation they’ve provided to make sure I fully understand what they want. My approach is very client-focused: I want them to be pleasantly surprised by the final result. After that, I load the audio files into Cubase and choose the first song of the album or EP to set the overall direction. Then I listen closely and apply processing that enhances the music’s emotional and physical impact."

He mostly collaborates with DIY and local artists, including Delachute, Navet Confit, Samian, Malaimé Soleil, Hansom Éli, Geneviève Racette, Barrdo, and Laura Niquay. Additionally, he has done remasters for Taima and Stefie Shock, among others.

"It's hard to choose a project I especially liked because I get to work on so much great music. Honestly, I'm really lucky: I would genuinely listen to most of what I work on."

"I'd say I especially love Delachute’s music. I was a fan even before he reached out to have me master his tracks, which felt like a real blessing. But I could name a long list of projects I truly enjoyed working on."

How LALAL.AI Is Integrated in the Music Mastering Workflow

Before using LALAL.AI, he tried tools like iZotope RX or sometimes no specialized software at all. He also tested other alternatives but found that their outputs didn’t perfectly match the untreated original when recombined.

"LALAL.AI is fast and easy; with it, I have more separation options than RX.

"What makes me trust LALAL.AI stem separation algorithm is that I made a test to make sure there is no artefact when I recombine. To be more specific, I take, for example, the separated version and play it together with the polarity-inverted original version of the full mix in sample-accurate synchronization. It’s what we call a 'null test.' If it’s complete silence, it means that there is no harm to the signal during the process.

"LALAL.AI is able to separate stems that can be recombined and sound exactly the same as the original source."

"That is really important for me as all processes in my chain have to be transparent. Or, at least, I have kept full control over the modifications I induce in the process."

Villemure says he's always looking for innovations in the de-noise and de-reverb technology, mostly for his podcast and audio post-production of the video content side of the business. The first time he ever tried LALAL.AI was to de-noise a dialogue that was hard to clean with his other tools.

"In a podcast that I had to clean, edit and mix, one speaker had a bracelet that made a clinking sound throughout the conversation. If I had to remove it manually with RX or other spectral options, it would have taken me very boring and unproductive hours…

"With LALAL.AI, I just had to put the entire track into the noise removal tool, and the separation was perfect. The client was happy, I saved hours and money."

Read also: "Almost all of our releases were successfully produced thanks to LALAL.AI": How a Label Uses AI to Fix the Unfixable

"The best use case for the de-noise algorithm I found that other options couldn't achieve is when a speaker in a podcast or live recording hits the microphone. LALAL.AI can separate the noise and the signal almost perfectly."

Villemure uses LALAL.AI primarily for de-noise purposes, but mastering tasks also have a place for the technology:

"For mastering tasks, I sometimes separate stems when I need some balance change of 1 dB or less between instruments. If it’s more than this, I typically ask the mixer for a modification. There won't be any audible artifact inside a 1 dB change on an AI-separated source.

"LALAL.AI gives me confidence that I have special tools that make me achieve some magic tricks that will impress my clients 🙂"

"My clients were impressed that a source could be cleaned when they thought they had to live with crappy sound! I always share my secrets, so I've already shared LALAL.AI on my Facebook page and to some open-minded colleagues. I see LALAL.AI as some pre-trained assistant specialized in different boring or redundant tasks."

Jean-Philippe isn't afraid of AI replacing human experts. He thinks that artists who "create music or high-quality, meaningful content want to work with people who are sensitive and thoughtful, and AI is just a tool to free up time for that."

Follow LALAL.AI on Instagram, Facebook, Twitter, TikTok, Reddit, LinkedIn, and YouTube to keep up with all our updates and special offers.