Tag: sound

The voice of computers croaks

sultan.sulf1James Flanagan, the man who helped computers to talk has died of a heart attack.  He was 89.

As a leading researcher at AT&T Bell Laboratories, Flanagan pioneered the developing field of acoustics and developed the tech for speech recognition, teleconferencing, MP3 music files and the more efficient digital transmission of human conversation.

He was famouse for his 1976 article, “Computers That Talk and Listen: Man-Machine Communication by Voice,” that appeared in Proceedings of the I.E.E.E., a journal published by the Institute of Electrical and Electronics Engineers.

More than 39 years ago, the point is that it painted a picture of society in the 21st century that actually happened.

Flanagan had his name on 50 patents. These included an artificial human larynx and a typewriter activated by the same audio tones as a push-button phone for the deaf.

His innovations included preserving the sound of a human voice while crunching it digitally.

He also taught computers to articulate by converting sound waves into digital pulses.

In 1974, Flanagan was one of six acoustical experts appointed by the courts in the President Nixon Watergate scandal. He proved that 18.5 minutes of a conversation between Nixon and his chief of staff, H. R. Haldeman, on June 20, 1972, were deleted in five separate erasures and re-recordings requiring “hand operation of keyboard controls”.

The conversation took place three days after the break-in at Democratic National Committee headquarters in the Watergate hotel and office complex in Washington.

Nixon’s personal secretary, Rose Mary Woods, accepted blame for erasing the first five minutes of the tape, saying she had been interrupted by a telephone call while transcribing it. Her explanation was dismissed as technically implausible. Nixon later resigned under threat of impeachment for, among other charges, withholding evidence.

Flanagan had come a long way from his family’s cotton farm in Greenwood, Mississippi Delta.

He joined the Army Air Forces at 17, worked to perfect signal scrambling and radar during World War II. Afterwards he returned to Mississippi State University and later received a master’s degree and a doctorate from M.I.T. In 1956, he joined Bell Labs, where he would work for 33 years. He retired in 1990 as director of information principles research.

In addition to his wife, Mildred, he is survived by his sons, Stephen, James and Aubrey and five grandchildren.


Kate Bush is the best way to deal with drones

97334cab7b3dc626b77b25cb6f686dacWho would have thought that Kate Bush would be the inspiration for knocking drones out of the sky>

Apparently some drones have a problem with resonance so if you conduct an experiment of sound on them you could cause them to crash.

Components inside drones are susceptible to certain pitches .

Researchers at the Korea Advanced Institute of Science and Technology (KAIST) in Daejon, South Korea, analysed the effects of resonance on a crucial component of a drone, its gyroscope.

A gyroscope keeps a drone balanced, providing information on its tilt, orientation and rotation, allowing for micro-adjustments that keep it aloft. Hobbyist and some commercial drones use inexpensive gyroscopes that are designed as integrated circuit packages.

Gyroscopes have been designed to have resonant frequencies that are above the audible spectrum, said Yongdae Kim, a professor in KAIST’s electrical engineering department. But others are still in the audible spectrum, making them vulnerable to interference from intentional sound noise.

Matching the resonant frequency of a gyroscope causes it to generate erroneous outputs, which have an effect on its flight, Kim said.

Researchers attached to a drone a small, consumer-grade speaker that was wirelessly connected to a nearby laptop.

The drone takes off normally, but when the right noise is played through the speaker, it smacks into the ground.

“When the gyroscope starts fluctuating, it affects the rotor speed directly,” Kim said.

At 140 decibels, it would be possible to affect a vulnerable drone up from around 40 metres away, Kim said.

There are a variety of sound-related offensive and defensive devices already on the market. For example, the LRAD Corporation makes the 450XL, which it terms an “acoustic hailing device.”

It can be mounted on a vehicle or a tripod and can project a voice message up to 1,700 meters so would be a good way of delivering an attack on a drone at close quarters.

Researchers use smartphone to create indoor map

A team of Swiss researchers has worked out a way to use a smartphone to create a map of the inside of a building.

According to the study’s lead author electrical engineer Ivan Dokmanic of the Swiss Federal Institute of Technology in Lausanne, an algorithm—based on the sort of echolocation that bats and dolphins use to navigate—could be incorporated into mobile phone apps to determine room dimensions for architectural or design purposes

He claims it could also be used to develop more realistic echoes in video games and virtual reality simulations and to eliminate the echo from phone calls.

Sorting out individual echoes from a recording is technically challenging and it needs microphones.

The researchers had someone stand in the centre of the room and snap their fingers or pop a balloon. Then they developed a mathematical algorithm to analyse the recordings from each microphone. Their method first eliminated echoes that had bounced off more than one wall or off small objects within the room in order to simplify the problem.

They worked out that every echo could be considered mathematically equivalent to a sound that emanated from a mirror image of the source.

This meant that they could manipulate the data that turned out to be the key to sifting through the sounds.

Dokmanic and his colleagues took the setup to Lausanne Cathedral where it was able to map the Gothic structure.

Integrating the calculations in other programs would open up real-world applications, Dokmanic says.

Mobile phone apps—requiring a few phones in one room—could be programmed to spit out room dimensions faster than it takes to pull out a tape measure and do it manually. 

PaloAlto's Cubik speakers reviewed

Here we have PaloAlto’s new laptop speakers which are arguably aimed at the Apple crowd. Compatible with both PC and Mac, PaloAlto says the Cubik speakers are a high end system that offers sound which is unmatched by other speakers in its class. 

The first thing you’ll probably notice about Palo Alto’s speakers is the design – a cube but at a weird angle – which makes fitting them on your desk kind of a pain if you live among clutter. I do.

But once you’ve figured out how to screw the base on (not hard) and plugged the things in, they complement a stylish laptop but look very out of place with a desktop. That was the idea. PaloAlto says that the way they are designed means you can place them anywhere in a room and get the same quality of sound – which you do. 

Unfortunately they are not particularly portable, so one assumes they are intended for the user with a desktop replacement laptop. Fortunately for Palo Alto there are a lot of those out there, and they’re available to buy on the Apple store, which shouldn’t hurt revenues. Again, the portability is testing for someone who moves around a lot but doesn’t like the  generally tinny sounds that come out of, say, a netbook. Headphones are still the best bet on that front. Especially because you’re going to need a power supply – these aren’t some flouncey USB powered speakers, they need proper juice.

Impressively, they handled Skream’s dub island and you could still hear the beats over the top. http://www.youtube.com/watch?v=6eWnp_RwKCU It struggled about half-way through, but a worn out or elderly sub can give you the kind of wobble you don’t want to hear. The Cubiks have long-vent advanced enclosure, which is the technology that enables the rich sound and bass from a small system. 

A problem was the controls. They’re on the speaker itself, which is fine, but the way the individual speaker is designed and sits on your desk means you’ve got to reach around to turn the volume up or down. The buttons were not particularly clearly marked, so if you’re a forever alone basement dweller they will be no good for 2am bouts of Starcraft with the lights off. 

You can tell why when you switch them on. The design is incredibly clever. It manages to deliver a respectable amount of bass without totally muddying the music. Of course, it won’t compare to a real sub-woofer or high end gear, but in a touch, they’re an OK buy for what they can do. 

PaloAlto claims that the Cubiks are of a comparable quality to BOSE PC speakers and for half the price, at £179.49. Admittedly, there are similar options in the same class that cost a lot more – but we would still recommend a full set up for sound buffs regardless. Those are upgradeable. I have to say that my six year old Creative I-Trigue speakers, which have been used an awful, awful lot, still deliver clearer clarity and overall sound quality, with a fuller experience. Personally, the price tag seems a little high for the product you get. Although it is undoubtedly a smart design with impressive quality for what they are, music buffs will still probably prefer a more serious, fuller option.

We had a dubstep DJ and music producer give the speakers a run through. He was impressed with the punch they pack for the size and set-up, but ultimately, said in a pinch they wouldn’t do for basic sound engineering or, in his opinion, listening. But he still liked them, because they are a likeable product.

Creative releases a quad core sound processor

Creative, better known for its Sound Blaster product range, has announced a multicore sound and voice processor which is targeted at those who want their noise with less power.

Dubbed the Sound Core3D, the chip is designed for embedded systems. It is supposed to be offering improved sound quality over existing on-board audio. For years a dedicated sound card has been seen as an improvement on anything that comes from on board audio but this is supposed to be the answer to the problem.

Singaporean Creative vice president Steve Erickson said his team designed a sound and voice processor that means OEM partners can deliver the highest quality Sound Blaster audio ever on a motherboard.

It also means that there will be a “new level” of quality to sound and voice processing to consumer electronics products. Noisy, we guess.

Sound Core3D is a versatile chip that can deliver the highest performance voice processing and audio playback from a single chip, he claimed.

The chip has multiple DSP cores and a high-quality HD audio codec. Armed with four independent processing cores, six-channel 24-bit 102dB digital to analogue converters, four-channel 24-bit 101dB analogue to digital converters, integrated headphone amplifier, digital microphone interface, S/PDIF inputs and outputs, the chip has all the general purpose inputs and outputs you can eat.

Word on the street is that Gigabyte will put the processor in its future products. 

Charge your phone by shouting at it

Pretty soon the bloke who is shouting “I am on the train” might simply be charging his phone, rather than being an annoying tosser who should be first against the wall when the revolution comes.

Boffins have worked out how to turn energy from sound into electricity and this might open the way for mobiles to be charged by actually talking.

South Korea’s Dr Sang-Woo Kim of Sungkyunkwan University in Seoul has developed a prototype of the technology that can convert sound of around 100 decibels into 50 millivolts of electricity.

While this is not enough to charge much, it is early days in the technology’s development.

What he did was shove zinc oxide wires between two electrodes and a pad on top vibrates when hit by sound waves. This causes the wires to compress and release, generating an electrical charge for small devices such as mobile phones.

If a sound is really loud it can generate more juice.

It is not just mobile phones either. If you built sound-insulating walls near motorways you could generate electricity from the sound of passing vehicles.

Still, if shouting at your phone generates electricity, the gunshots of commuters turning militant on the train bloke for charging his phone should generate a few millivolts too. 

Mitsubishi develops carbon nanotube high-end speakers

Mitsibushi Electrics has developed a prototype speaker with carbon nanotube-based cones, which are said to offer the same audio quality as existing high-end materials but at a lower cost.

According to nikkei.com, the company hopes that the cones will set the ball rolling for less expensive speakers. It hopes to push these into the public domain for use in TVs and cars.

The speakers works by incorporating plastics and nanotubes, which are said to give the same sound quality as those which use metal and ceramics and are more expensive.  According to the company they also deliver high-quality sounds in a wide range of pitches.

The company’s high-end speakers, which cost several hundreds of thousands of yen, use cones made of boron carbide. And because they are produced in a vacuum unit, the process can become pricey.The carbon nanotube-based cones, however, can be manufactured at lower cost as the new materials can be incorporated into a plastic injection molding, which is comparable to a high-velocity metal diaphragm.

The company said the development of new materials and the optimum formulation of resin carbon several would make these speakers succeed. It said it’s new speakers were comparable to the widespread use of titanium metal diaphragm velocity higher than 5000m.

It also said these new speakers were closer to the source of sound reproduction unity thus accurately being able to reproduce the sound. They also have a woofer speaker vibration plate material for the equalisation of high notes in songs.

Scientists work out why mobiles annoy

Top scientists have been working out why mobile phones wind you up so much.

Cornell University’s finest minds know that people shouting into mobile phones get you cross but wondered exactly why.

According to the journal Psychological Science, which we get for the spot the neurosis competition, the boffins claim that it is all because of the way that the mind works.

Apparently we have less control to move away our attention from half a conversation than we do when we are listening to a complete one.

“Since halfalogues really are more distracting and you can’t tune them out, this could explain why people are irritated,” she said in an interview.

Emberson said that the brain half hears the conversation and tries to make sense of it.

This is because part of the brain is dedicated to predicting what speakers will say next.

“When you hear half of a conversation, you get less information and you can’t predict as well,” she said. “It requires more attention.”

The findings by Emberson and her co-author Michael Goldstein were based on research involving 41 college students.

They did concentration exercises, like tracking moving dots, while hearing one or both parties during a mobile phone conversation.

The students made more errors when they heard one speaker’s side of the conversation than when they overheard the entire dialogue.

Boffins discover the sound of silence

A team of Oregon boffins has identified a part of the brain that deals specifically with shutting off sound processing.

They say a bit of silence is vital for hearing and for understanding speech.

Writing in the journal Neuron, which we get for the “spot the brain cell” competition the discovery flies in the face of a long-held assumption that the signalling of a sound’s appearance and its subsequent disappearance are both handled by the same neural pathway.

The new finding, which supports an emerging theory that a separate set of synapses is responsible, could lead to new, distinctly targeted therapies such as improved hearing devices.

Michael Wehr, a professor of psychology and member of the UO Institute of Neuroscience said that it looks like there is a whole separate channel that goes all the way from the ear up to the brain that is specialized to process sound offsets.

The two channels come together in a brain region called the auditory cortex, situated in the temporal lobe.

Wehr, Ben Scholl, and Xiang Gao monitored the activity of neurons and their connecting synapses as rats were exposed to millisecond bursts of tones, looking at the responses to both the start and end of a sound. They tested varying lengths and frequencies of sounds in a series of experiments.

One set of synapses responded “very strongly at the onset of sounds,” but a different set of synapses responded to the sudden disappearance of sounds.

Being able to perceive when sound stops is very important for speech processing. This discovery answers the question about how the brain find the boundaries between the different parts of words, particularly when you are in a noisy pub.