1aSC31 – Shape changing artificial ear inspired by bats enriches speech signals – Anupam K Gupta

1aSC31 – Shape changing artificial ear inspired by bats enriches speech signals – Anupam K Gupta

Shape changing artificial ear inspired by bats enriches speech signals

Anupam K Gupta1,2 , Jin-Ping Han ,2, Philip Caspers1, Xiaodong Cui2, Rolf Müller1

1 Dept. of Mechanical Engineering, Virginia Tech, Blacksburg, VA, USA
2 IBM T. J. Watson Research Center, Yorktown, NY, USA

Contact: Jin-Ping Han – hanjp@us.ibm.com

Popular version of paper 1aSC31, “Horseshoe bat inspired reception dynamics embed dynamic features into speech signals.”
Presented Monday morning, Novemeber 28, 2016
172nd ASA Meeting, Honolulu

 

Have you ever had difficulty understanding what someone was saying to you while walking down a busy big city street, or in a crowded restaurant? Even if that person was right next to you? Words can become difficult to make out when they get jumbled with the ambient noise – cars honking, other voices – making it hard for our ears to pick up what we want to hear. But this is not so for bats. Their ears can move and change shape to precisely pick out specific sounds in their environment.

This biosonar capability inspired our artificial ear research and improving the accuracy of automatic speech recognition (ASR) systems and speaker localization. We asked if could we enrich a speech signal with direction-dependent, dynamic features by using bat-inspired reception dynamics?

Horseshoe bats, for example, are found throughout Africa, Europe and Asia, and so-named for the shape of their noses, can change the shape of their outer ears to help extract additional information about the environment from incoming ultrasonic echoes. Their sophisticated biosonar systems emit ultrasonic pulses and listen to the incoming echoes that reflect back after hitting surrounding objects by changing their ear shape (something other mammals cannot do). This allows them to learn about the environment, helping them navigate and hunt in their home of dense forests.

While probing the environment, horseshoe bats change their ear shape to modulate the incoming echoes, increasing the information content embedded in the echoes. We believe that this shape change is one of the reasons bats’ sonar exhibit such high performance compared to technical sonar systems of similar size.

To test this, we first built a robotic bat head that mimics the ear shape changes we observed in horseshoe bats.

 

Figure 1: Horseshoe bat inspired robotic set-up used to record speech signal

han1

 

We then recorded speech signals to explore if using shape change, inspired by the bats, could embed direction-dependent dynamic features into speech signals. The potential applications of this could range from improving hearing aid accuracy to helping a machine more-accurately hear – and learn from – sounds in real-world environments.
We compiled a digital dataset of 11 US English speakers from open source speech collections provided by Carnegie Mellon University. The human acoustic utterances were shifted to the ultrasonic domain so our robot could understand and play back the sounds into microphones, while the biomimetic bat head actively moved its ears. The signals at the base of the ears were then translated back to the speech domain to extract the original signal.
This pilot study, performed at IBM Research in collaboration with Virginia Tech, showed that the ear shape change was, in fact, able to significantly modulate the signal and concluded that these changes, like in horseshoe bats, embed dynamic patterns into speech signals.

The dynamically enriched data we explored improved the accuracy of speech recognition. Compared to a traditional system for hearing and recognizing speech in noisy environments, adding structural movement to a complex outer shape surrounding a microphone, mimicking an ear, significantly improved its performance and access to directional information. In the future, this might improve performance in devices operating in difficult hearing scenarios like a busy street in a metropolitan center.

 

Figure 2: Example of speech signal recorded without and with the dynamic ear. Top row: speech signal without the dynamic ear, Bottom row: speech signal with the dynamic ear

han2

 

3pAB2 – A design for a biomiometic dynamic sonar head. – Phllip Caspers, Yanqing Fu, and Rolf Müller

WASHINGTON, D.C., May 20, 2015 — Engineers at Virginia Tech have taken the first steps toward building a novel dynamic sonar system inspired by horseshoe bats that could be more efficient and take up less space than current man-made sonar arrays. They are presenting a prototype of their “dynamic biomimetic sonar” at the 169th Meeting of the Acoustical Society of America in Pittsburg, Penn.

Bats use biological sonar, called echolocation, to navigate and hunt, and horseshoe bats are especially skilled at using sounds to sense their environment. “Not all bats are equal when it comes to biosonar,” said Rolf Müller, a mechanical engineer at Virginia Tech. “Horseshoe bats hunt in very dense forests, and they are able to navigate and capture prey without bumping into anything. In general, they are able to cope with difficult sonar sensing environments much better than we currently can.”

To uncover the secrets behind the animal’s abilities, Müller and his team studied the ears and noses of bats in the laboratory. Using the same motion-capture technology used in Hollywood films, the team revealed that the bats rapidly deform their outer ear shapes to filter sounds according to frequency and direction and to suit different sensing tasks.

“They can switch between different ear configurations in only a tenth of a second – three times faster than a person can blink their eyes,” said Philip Caspers, a graduate student in Müller’s lab.

Unlike bat species that employ a less sophisticated sonar system, horseshoe bats emit ultrasound squeaks through their noses rather than their mouths. Using laser-Doppler measurements that detect velocity, the team showed that the noses of horseshoe bats also deform during echolocation–much like a megaphone whose walls are moving as the sound comes out.

The team has now applied the insights they’ve gathered about horseshoe bat echolocation to develop a robotic sonar system. The team’s sonar system incorporates two receiving channels and one emitting channel that are able to replicate some of the key motions in the bat’s ears and nose. For comparison, modern naval sonar arrays can have receivers that measure several meters across and many hundreds of separate receiving elements for detecting incoming signals.

By reducing the number of elements in their prototype, the team hopes to create small, efficient sonar systems that use less power and computing resources than current arrays. “Instead of getting one huge signal and letting a supercomputer churn away at it, we want to focus on getting the right signal,” Müller said.