Nokia Explains How A Camera’s AI “Can Tune the Audio to Fit the Scene”

Although AI took a step back at Mobile World Congress Shanghai, letting 5G and biometrics come forward, the technology came up in various discussions on the floor. One of them was with Paul Melin, VP of Digital Media at Nokia, who told us that AI will play a big role in mobile audio. In fact, their software allows OEMs to use phone cameras to tune it according to the scene captured. #mwcs18 #ttllive

As you know, Nokia OZO was the last big product coming from the Nokia team. As the mobile business was taken by HMD Global, Nokia kept the OZO project for a while. Unfortunately, the disappointing profit it made as well as the lay-offs forced Nokia to stop the production of the spherical camera. 

What remains is the spatial audio, the software Nokia offers now to OEMs. Paul Melin believes that this is their ace in the sleeve right now; although the video recording capabilities of devices have seen improvements, most phones still offer only mono audio. Nokia has therefore the opportunity to bring spatial audio on flagship phones without requiring hardware and design adjustments.

“Our software is so flexible that it can tune our algorithms to the microphone placements that the device manufacturer has in their device”, explained Melin. You can zoom and focus on a particular person in a scene where people talk.

Even better, those phones that are equipped with machine learning can adjust the audio. The artificial intelligence selects the relevant sounds in a clip, independent of the scene. Sounds complicated? It’s really not! You won’t have to do anything, just let the software make decisions after recognizing the scene captured.

It was about time audio caught up with video, right? Watch the video above to find out in which devices Nokia OZO pop up!