New Tech Silences People With Own Words

Two Japanese scientists have designed a device that can disorient a talking person powerfully enough to make them stop. The technique is based on the "echo" feedback we sometimes get when using a telephone.

Kazutaka Kurihara and Koji Tsukada created the SpeechJammer while investigating the characteristics of speech control.

They were particularly interested in "problem" scenarios where loud talkers annoyed unwilling listeners, and where conversations hit a snag because both speakers continued to talk at the same time.

For example, if somebody is speaking in a way that annoys others (as in a library, for example), it's often difficult to simply tell them to stop speaking.

You may have to wait to get a word in edgeways, shout so loudly you overpower them, get their attention with a visual message, or even use physical contact.

As the pair put it, "additional abuses of unavoidability and occupancy... may occur and could lead to further conflict." (Source: arxiv.org)

Echo Effect Throws Speaker Off Course

Their solution emulates the way telephones sometimes give us unwanted feedback: when a speaker hears his or her own words just after they've spoken them, it often creates an awkward feeling.

In some cases such an "echo" is even enough to cause people who normally speak clearly to develop a temporary stammer.

Kurihara and Tsukada used this idea to design a gadget that records speech using a directional microphone, then plays it back on a delay, beaming the sound back towards the person speaking.

Device Uses Laser Pointer, Eight AA Batteries

The device, which includes a laser pointer so the user can be sure of accurately targeting the talker, is small enough to be held in one hand.

It's powered by eight AA batteries, which isn't particularly convenient or efficient, but is necessary to avoid possible noise interference from power cables.

However, despite its small size and battery power, the gadget isn't quite portable (yet); it needs to be hooked up to a PC that can process the sound and control the delay.

Variable Delay Means Brain Can't Cope

Testing of the device found that, as a general rule, the longer the delay the greater the effect. However, the inventors say they need to test a wider range of delays to find the point, if any, at which this increased delay stops increasing the device's effectiveness.

The pair also discovered that varying the delay is more effective than sticking to a fixed delay, presumably because it limited the ability of the brain to compensate for the interference.

The system works better when the talker is using a consistent tone and volume, such as reading out aloud, than it does for conversational speech.

Perhaps unsurprisingly, it didn't work at all when the speaker was simply repeating a meaningless sound over and over. (Source: cnn.com)

There's no suggestion of this becoming a commercial product, and at this stage it's purely a case of seeing what can be learned from it.

However, Kurihara and Tsukada suggest their discovery could have practical uses, either to enforce requests for quiet (such as in a library), or to enforce a rule that only one person speak at a time in a meeting.