Microsoft Wants to Censor Xbox Live Voice Chat in Real Time
Could a nimble AI beep out offensive phrases in real time?
Online gaming communities are rife with toxic behavior — sign on to Xbox Live and it’s a near-guarantee that you’ll hear some of your fellow gamers hurling insults and slurs as readily as they lob grenades in “Call of Duty.”
“It’s regular, every other game you’re in, there’s always someone who has a mic or types in chat,” then-16-year-old gamer Bailey Mitchell told the BBC in 2017. “They’ll call you some random abusive thing they can think of.”
In an effort to fight back against that toxicity, Microsoft announced Monday that it had begun rolling out new filters for Xbox Live designed to block out potentially offensive messages — and in a futuristic twist, the company is already trying to figure out how to do the same for voice messages.
According to a Microsoft blog post, Xbox users will be able to choose between four levels of content filtration: Friendly, Medium, Mature, and Unfiltered.
If the filtration system flags a message as being beyond the limits the player has set, it will replace the message with a placeholder reading “potentially offensive message hidden.” The player can then click on the placeholder if they want to read the message anyway.
But filtering text-based messages is just the first step.
Microsoft Research is already trying to crack real-time speech-to-text translation — a technology that would allow it to transcribe a verbal conversation essentially as it takes place — and the company’s Xbox division is already contemplating what that could mean for its online community.
“What we’ve started to experiment with is ‘Hey, if we’re real-time translating speech to text, and we’ve got these text filtering capabilities, what can we do in terms of blocking possible communications in a voice setting?'” Dave McCarthy, head of Microsoft’s Xbox operations, told The Verge.
Rob Smith, a program manager on the Xbox Live engineering team, added that the “ultimate goal” would be a system that could “detect a bad phrase [in voice conversations] and beep it out for users who don’t want to see that.”
He compares it to broadcast TV, though television censors have the benefit of a seven-second delay to aid their filtration of offensive content — delaying voice communications between gamers for even a fraction of a second could dramatically impact game play.
Still, it’s easy to see how transcription technology is already moving in the direction Microsoft envisions, so while gaming communities might currently be cesspools of toxicity, that might not always be the case.
“If we really are to realize our potential as an industry and have this wonderful medium come to everybody, there’s just no place for that,” McCarthy told The Verge.
READ MORE: Microsoft Unveils Xbox Content Filters to Stop the Swears and Toxicity [The Verge]
More on Microsoft: Microsoft: We Stopped Listening to Your Xbox Convos “Months Ago”