Introducing the next generation of audio processors

The buzz around AI tools can, at times, appear deafening. It seems like every company out there is claiming to have AI integrated into their products. We’re all familiar with AI assistants on our phones that can help with tasks like writing emails, summarizing meetings, and translating text. But what about AI-powered washing machines? AI-powered toasters? There are even AI-powered pet feeders. Amidst all of the hype, it can be hard to identify those tools that can genuinely provide meaningful solutions from those that are, well, a bit gimmicky.

AI-generated...everything?

When talking about audio and AI, two things that probably come to most people’s minds are AI-generated music and deepfake audio. The former definitely falls into the gimmicky category (John Lennon and Mozart will not be quaking in their boots just yet), while the latter crosses all sorts of ethical boundaries and is probably not a good role model for these new technologies.

Moving Beyond The Gimmicks

What Media Organizations really need is not some little tool that can render a pseudo-Beatles composition or mimic some famous person, but something that will truly add value to their operations. This is where we at Telos Alliance see the AI direction of travel going. 

Advances in technology, particularly those driven by artificial intelligence or machine learning, allow the creation of audio processors and analysis tools that one could have only dreamed of previously. Tools such as language detection, voice separation, and dialog intelligibility analysis can potentially revolutionize audio workflows. And the perfect home for these new technologies? Why, AudioTools Server of course!

Putting AI to Work The AudioTools Server Way

Where AudioTools Server has the ability to take AI-driven processes to the next level is its ability to integrate them as part of a wider audio workflow. Let’s take language detection as an example. Being able to analyze a file and identify the likely main language is one thing (and pretty impressive, I should add), but the power really comes in when you can take that information and do something with it. 

Imagine you have a workflow that requires you to deliver files with English on the first two channels and Spanish on the next two channels. As a large organization, you receive hundreds of files every day from a variety of different suppliers, including archive content. In the past, the only way to reliably guarantee that you are delivering the correct audio would be to manually QC the files. With an AudioTools Server workflow the files can be analyzed, their language identified and then the results used to either pass the file on (if correct), reject it outright (if one or more of the required languages could not be identified) or (and here’s the exciting bit) automatically re-order the channels on the file so that the English and Spanish audio is exactly where you need it to be.

Another example would be identifying and improving sections of audio where there are potential problems with dialog intelligibility. While tools for assessing the overall dialog level and comparing it with the overall program loudness have been around for many years, new analysis tools take this further, providing an intelligibility rating based on how the dialog would actually be perceived by somebody listening to the audio. In an AudioTools Server workflow, the analysis results could then be used to trigger another AI-powered function, dialog extraction, which separates the dialog and non-dialog audio components into separate files. The extracted dialog could then be adjusted, correcting the problem areas, before mixing back with the rest of the audio to create a more intelligible mix.

AudioTools Server helps make workflows smarter.

Smarter Audio, One Slice at a Time

These are just a few examples of how the next generation of audio processors, when combined with the powerful audio workflow engine of AudioTools Server, can create intelligent, fully automated workflows that solve challenges that previously could only be solved through manual processes.

So forget about all of the gimmicky things out there that claim to be powered by AI and think about how, in the correct hands, these new technologies can help save you time and money, opening up avenues that were previously unimaginable. That said, I do quite like the sound of the AI-powered toaster. Now, where did I leave the honey?

To learn more about how AudioTools Server can help your audio workflows flow smarter, contact your AudioTools Server representative. We’d love to hear from you.