I saw the LG C5 OLED TV’s new personalized sound mode in action, and it’s the best AI TV feature I’ve seen so far
A couple of years ago, LG introduced an AI-powered Personalized Picture Wizard mode in its TVs, with an interesting take on the feature: instead of the traditional TV thing of making you choose between modes with not-especially-descriptive names like ‘Movie’ or ‘Standard’, you’re shown multiple different versions of the same images, you choose which ones appeal to you most, and then a picture preset is created for you using AI algorithms, based on what you chose.
The one problem with this? We hated the results. It was straight back to Filmmaker or a Dolby Vision mode for us. The problem is that while there’s certainly a wide variety of subjective preferences for how things look, we all know roughly what looks ‘correct’ for movies and TV, and there’s not a big reason to deviate from that – unless you’re choosing something to make the screen more visible to people with vision problems, although accessibility modes weren’t part of what this wizard offered.
Now, in 2025, LG is adding a Personalized Sound Wizard mode to its TVs, and I think this is a much better use of this kind of tech. People’s hearing is a lot more nuanced than vision – different people’s hearing can be weaker or stronger in particular frequency areas, some people struggle to pick dialogue out of a complex sound mix more than others for neurological rather than ear-based reasons, and there are people with the beginnings of hearing loss who don’t know it yet, or are resistant to hearing aids, and all of these people be helped by targeted audio tweaks.
You may like
That’s what the Sound Wizard can offer. Like the Picture version, it’ll give you four different versions of the same audio clip, and you move the selection over each one to hear what it sounds like. You choose the one or two that you find most preferable, and then move onto a new audio clip. Based on your feedback, it’ll create a sound profile for you that, for example, emphasizes dialogue and bright sounds, or adds more fullness to lower registers.
At the end, it’ll tell you what kind of sound it thinks suits you, and will play you a demo during which you can switch between the personalized mode and the default mode, so you can see if you actually like what it’s offering.
The process is super-simple, and once it’s done, that sound profile is tied to your user account on the LG TV, meaning that if someone else uses the TV with their account, they can have a totally different sound profile (another custom one, or one of the presets).
TVs that have the feature also let you use your voice to switch account – if you use voice commands to ask to watch something or open a particular service, it’ll know who’s asking, and will switch accounts and presets accordingly.
The difference that profiles it creates seems to be quite significant, though I was shown in a demo space with limited testing ability, and wouldn’t be able to make a full judgment until we really get into it during a full review.
But I’m fully on board with the principles here, and I think this is the kind of use case where AI machine learning is a genuinely useful thing to add to TVs – it ranks up there with some other features I’ve seen, such as as auto-generated subtitles for shows that don’t have them, or auto-translation of shows in another language; tools that help us to hear, see or understand things that we couldn’t before.
This feature will be available on LG’s webOS 25 TVs, including the LG C5, LG G5, LG B5, and LG’s new and improved QNED range. LG will bring webOS 25 to older TVs too, starting with 2024 TVs probably late in 2025 or in early 2026. Whether this feature will make it over isn’t clear yet, but I hope it comes to as many people as possible.