New Teams features use AI and machine learning to improve video and audio experiences

On June 13, Microsoft released numerous new features to its communication platform Teams, focusing on the improvement of its voice and video quality. In particular, these new Teams features include echo cancellation, “de-reverberation,” background noise suppression, “interruptibility” capability, real-time screen optimization, AI-based optimization, and brightness and focus filters.

“Disruptive echo effects, poor room acoustics, and choppy video are some common issues that hinder the effectiveness of online calls and meetings,” Microsoft Teams Vice President Nicole Herskowitz says, addressing the main concerns that pushed the development of the new Teams features. “Through AI and machine learning, which have become fundamental to our strategy for continual improvement, we’ve identified and are now delivering innovative enhancements in Microsoft Teams that improve such audio and video challenges in ways that are both user-friendly and scalable across environments.”

The new echo cancellation capability of Teams isn’t the only feature focusing on the echo problems being encountered by the users when communicating within the platform. Some of the audio features introduced by Microsoft also target other echo-related issues caused by the environment and audio devices. For instance, the new de-reverberation feature utilizes a machine learning model that can convert captured audio signals to sound, preventing the sound from reverberating or the user from sounding like speaking inside a cave.

Another feature also resolves the echo-related problem caused by audio devices placed near each other. “A natural element of conversation is the ability to interrupt for clarification or validation,” says Herskowitz. “This is accomplished through full-duplex (two-way) transmission of audio, allowing users to speak and hear others at the same time. When not using a headset, and especially when using devices where the speaker and microphone are very close to each other, it is difficult to remove echo while maintaining full-duplex audio. Microsoft Teams uses a model “trained” with 30,000 hours of speech samples to retain desired voices while suppressing unwanted audio signals resulting in more fluid dialogue.”

According to Herskowitz, with the new interruptibility capability of the Teams, users can make smooth exchanges without feeling interrupted or cut out, which is commonly encountered on online communication platforms. In addition, the company announced making its machine learning-based noise suppression tech default for Teams (Windows, Mac, and iOS and soon to Teams Android and web clients), allowing clearer calls free from unwanted noises.

Microsoft also extends the use of AI to the video performance of Teams in order to create relevant visual adjustments, especially in cases where users encounter problematic bandwidth conditions when sharing content during presentations. For example, Teams’ real-time screen optimization feature that uses machine learning can detect and adjust content characteristics, resulting in optimized document legibility and smooth video playback. Meanwhile, by adjusting the video playback, the AI-based optimization prevents the drastic effects of choppy videos caused by poor network bandwidth. Lastly, the new AI-powered filters in Teams allow users to adjust brightness and use soft focus when they’re having meetings in low-light areas.

According to Microsoft, the said features mentioned are currently on the rollout and will be generally available in the coming months.