>
Microsoft Teams says it will end communication problems caused by sudden and temporary interruptions in internet connection with a new artificial intelligence-based solution for its video conferencing software.
In a demonstration on the Microsoft Teams blog (opens in new tab)Microsoft said its new machine learning model is built on an existing technique it calls “Packet Loss Concealment” (PLC), which appears to make “assumptions about the missing content” when a connection drops and a caller is speaking.
It also claimed that the AI model can predict “up to 80 consecutive milliseconds of audio,” a significant improvement over PLC implementations that are not enhanced by AI. This, it says, makes packet loss undetectable to users.
“Undetectable” drops in quality
That’s a bold claim, and while the early post-release feedback is positive (users report detecting packet loss 15% less), and the demonstrations are certainly listenable, there’s always room for improvement.
That improvement may come sooner than expected, as Microsoft has shown a willingness to share this technology with researchers and developers around the world.
For an event on conference on speech technology INTERSPEECH (opens in new tab) earlier this month, Microsoft released “open-source network traces” [it] collected”, and a “PLC Mean Opinion Score Model”, so that others could incorporate an existing dataset, built from over 600 hours of audio analysis, into their own work.
We shouldn’t expect miracles – Microsoft itself noted that it can’t improve the stability of a remote network – but this should alleviate the frustration of sudden disconnections, which the world of hybrid work since the start of the Covid-19 pandemic.
It should also mean that improved PLC implementations will become more common on the web in the future, as users wanting to see call quality improvements aren’t locked into Microsoft Teams.
In the meantime, however, Microsoft Teams users on Windows can now enjoy the improvements, which are designed to run only when needed to reduce CPU load. Microsoft has stated that Mac and mobile implementation are also underway.