Partner POV | All About AI on Logitech Devices

Article written by Craig Ishii, Solutions Marketing Lead, Logitech. 

There are three key elements to ensuring productive and collaborative meetings: being seen, being heard, and the ability to collaborate effortlessly. In this article, we'll explore Logitech's portfolio of AI-driven features with these principles in mind, enabling the best meeting experiences for employees.

See and be seen with RightSight 2

We'll start with the concept of 'see and be seen.' While any conference camera can provide a wide view of the room, we seek to bring the individuals in the room closer to those who are remote. Everything from facial expressions, to gestures and body language help to add crucial context to conversations. With Logitech's RightSight 2 AI features, we bridge the gap between remote and in-room participants. This technology captures subtle visual cues, adding depth and context to meetings and bringing participants closer together. 

Starting with Group View, the AI in our video bars use object detection to find people in the room, then zoom and center the group in the middle of the frame. Speaker View hones in on individuals as they speak using direction of arrival technology to pinpoint the location of voices and align them with detected individuals. Grid View takes this further by framing individuals separately, giving each person their own gallery image. Our AI also levels head heights and standardizes the zoom on each individual, ensuring everyone appears equal. 

Additionally, features like Camera Zone use on-device AI to determine individuals' distances from the camera. This allows for our video bars to identify who's inside or outside of the administrator-set zone not just on the right and left, but also with depth. 

As more cameras are added to the room, Logitech employs AI within each additional camera to enhance the meeting experience in multi-camera setups. Starting with Logitech Scribe, users first define the whiteboard boundaries enabling Scribe to ensure that broadcasted content is correctly proportioned. We then use color and sticky note enhancement to help remote participants clearly see what's written and posted on the board. Presenter transparency minimizes the blockage of content behind the person writing on the board. 

When Logitech Sight is added to the center of the table, it utilizes AI to depict individuals at similar head heights and proportions using the same technology mentioned previously. With the introduction of Smart Switching, Logitech Sight and the front-of-room Rally Bar communicate to agree upon the position of individuals in the room, then intelligently switch the feed based on the direction they're facing.

Our AI seamlessly integrates with Microsoft Teams, Zoom, and Google Meet, offering dynamic features that enhance the meeting experience. In Microsoft Teams, RightSight 2 powers Intelliframe, creating individual tiles for each participant to improve person-to-person interactions. In Zoom, our multi-stream technology delivers multiple video feeds, providing individual tiles for in-room participants while our high-quality visuals support Zoom's Smart Name Tags, making it easy for remote attendees to identify who's in the room. In Google Meet, the clear visuals from our video bars enable Dynamic Tiles, displaying up to three in-room participants in separate tiles for a more personalized view.

Hear and be heard with RightSound 2

Being heard is arguably more critical than being seen. If you can't hear what's being said, the meeting loses its value. Our RightSound 2 algorithms ensure clear, crisp audio by eliminating distracting noises, and creating an unrivaled voice clarity. 

Starting with AI Noise Suppression, we identify sounds that are not voices every 1/100 of a second, then use machine learning and spectral subtraction to eliminate the non-vocal sounds. Our devices also do Echo Suppression by utilizing AI to identify redundant vocal patterns coming from hard surfaces like floors, ceilings, and walls and eliminating those as well. 

After isolating voices, our device's AI uses Voice Equalization to level out the volume, ensuring all voices are broadcasted at similar levels. In rooms with multiple microphones, our devices identify and broadcast the clearest voice. 

Finally, in scenarios where there are Rally Mic Pods or the microphone on Logitech Sight in use, our Extended Audio technology identifies duplicate voices and transmits the voice from the nearest microphone ensuring the the clearest version is broadcasted during the meeting.

Better collaboration with productivity tools

Logitech's advanced audio and visual technology ensures that meeting content is captured accurately, no matter the platform. With Microsoft Teams' speaker recognition, clear audio from Logitech devices enables precise meeting summaries, takeaways, and transcripts. Copilot further enhances productivity by providing key insights, allowing employees to revisit meetings or catch up on what they missed.

In Zoom meetings, Logitech's RightSight 2 and RightSound 2 technologies power Zoom's AI-Companion which helps employees to review meetings and understand key takeaways.

Google Gemini will also leverage Logitech's clear audio and video to generate meeting summaries, transcripts, and even live translations, enhancing collaboration for global teams.

For IT and workplace service teams, Sync Insights uses data from our video bars to automatically track room and device usage, empowering them to make informed decisions about workspace management and real estate.

Ultimately, it's about productivity. With Logitech's suite of AI features in RightSight 2 and RightSound 2 delivering clear visuals and crisp audio, employees can focus on collaboration without distractions.