Automatic Censoring
[Computer][Future Technology]

Microsoft has filed a patent for automatic censoring of audio speech data. The idea is that a computer program will analyze the speech and try to break the speech down into individual phonemes, and then tries to match this against a database of undesirable sequences of phonemes.

The automatic censoring filter employs a lattice comprising either phonemes and/or words derived from phonemes for comparison against corresponding phonemes or words included in undesired speech data. If the probability that a phoneme or word in the input audio data stream matches a corresponding phoneme or word in the undesired speech data is greater than a probability threshold, the input audio data stream is altered so that the undesired word or a phrase comprising a plurality of such words is unintelligible or inaudible.

Analysis suspects that Microsoft's main motivation for research in this area is their XBox Live service. For those of you who don't know what that is, when you buy an XBox or an XBox 360, you can also purchase a subscription to "XBox Live", which allows you to play the game with other people across the Internet. Included as part of the kit you get when you subscribe is a headset allowing you to speak to your allies and/or competitors. Unfortunately, the general experience people have had with this service was encountering teenages who would swear and scream nonstop into their headsets, so usually people only enabled the voice system when playing with their friends.

With this automatic censor system, Microsoft could block certain sequences of phenomes (probably those describing "shit" and "fuck" and derived words like "fucked", "fucking", "fack", "shat", but not "shot" or "shoot"). Presumably, people could optionally turn on the live censoring when playing with strangers, and turn it off again when playing with friends.

E-mail this story to a friend.
, , , , , , ...

You must be logged in to post comments.

Sites linking to this post: