Interview with Tony Spahr, PhD, Principal Research Scientist, Advanced Bionics

CAROLYN SMAKA: Tony, can you tell me about your background and your role at Advanced Bionics?

TONY SPAHR: I have a Ph.D. from Arizona State University. I worked under Michael Dorman doing cochlear implant research at Arizona State University for several years after I graduated.  I’ve had the good fortune to be involved with cochlear implants for over 12 years. Earlier this year, I accepted a position with Advanced Bionics working in the research and technology group.

SMAKA:  What has it been like moving from a university to the industry side of cochlear implants?

SPAHR: It’s very exciting to be at Advanced Bionics – I’ve known many of the people here for a very long time and have a great deal of respect for what they have accomplished over the years.  So, for me, this was a very comfortable and enjoyable transition.  The idea of being in a position to affect meaningful change in the way cochlear implants are designed and to have a role in helping so many people hear was very appealing to me.  Moving into a worldwide organization and seeing all that goes into manufacturing a product has been quite a learning experience. It’s very compelling to see how things work when you’re inside of a company like Advanced Bionics, with the technological capabilities of the device and a large team of passionate, talented colleagues.  I’ve had the opportunity to explore new areas and to seek out innovative solutions for problems facing cochlear implant users, and so far it’s been a fantastic experience.

SMAKA: I’m excited to hear about the research on ClearVoice, from your perspective as an esteemed cochlear implant scientist.  Before we get in to the data, would you provide an overview of ClearVoice?

SPAHR: ClearVoice is a speech enhancement algorithm that was developed here at Advanced Bionics to improve understanding in real-world environments.  In most social situations, the sound picked up by a cochlear implant microphone will have a mixture of both speech and noise.  ClearVoice is designed to identify the noise in the signal and then attenuate it. The algorithm draws out the speech from the noise and allows the recipient to be a more active participant in a conversation. 

As you know, understanding speech in noise is a challenge for anyone with hearing loss.  As a researcher, we typically describe this challenge in terms of a change in performance on a speech test, such as the AzBio, HINT or QuickSIN.  However, for the listener, it’s much more than that. We’re talking about the ability to hear in real world environments.  People with hearing loss will completely alter their lifestyles because of the stress that they experience in noisy environments.  The effort that it takes for them to communicate in those situations can be overwhelming and exhausting, so much so that they’ll actively avoid noisy social situations. 

In the lab, we’ve proven that ClearVoice provides significantly better understanding of speech in noise.  Beyond that, recipients report that listening in real world noisy environments is easier.  Situations that were stressful and exhausting before they had ClearVoice are now not only tolerable, but even enjoyable. People are reporting that ClearVoice has literally changed the way that they’re living their lives; they’re more active and they’re getting more out of their social activities.

SMAKA: How exactly does ClearVoice work?

SPAHR: ClearVoice is proprietary technology from Advanced Bionics and it is built on our HiRes Fidelity120 sound processing platform. With ClearVoice, the signal that’s picked up from the microphone goes through the regular speech coding process.  In that process, the signal is broken down into a number of small frequency bands.  ClearVoice looks at the content of each one of these channels, analyzes it over time, and makes an estimate about how much noise is in a given channel.  It also looks for other signals in that channel, such as speech. As it builds up an understanding of the environment, it makes real-time decisions in each channel about whether the current segment is primarily speech or noise.  If it’s noise, ClearVoice attenuates the channel. It also makes decisions about the degree to which the noise should be attenuated based on the estimated signal-to-noise ratio in the environment.

ClearVoice is the only speech enhancement algorithm that is clinically proven to improve speech understanding in noise.  We have evidence from a clinical trial that was conducted at some of the most prestigious and respected medical clinics and universities in the country showing the benefits of ClearVoice.  They found that ClearVoice improved speech understanding in noisy environments, without compromising performance in quiet situations.  And, after patients finished the trial, 9 out of 10 said they would like to use ClearVoice in their everyday listening program, which I think speaks volumes.

SMAKA: Does ClearVoice have any effect in quiet?

SPAHR: The goal for a speech enhancement algorithm is to help understanding in noisy environments, but in order to do that you have to make changes to the incoming signal.  So, it is reasonable to be concerned about losing meaningful parts of the speech signal when listening in quiet.  However, it has been shown that, with ClearVoice, performance stays the same in quiet and is improved in noise.  For the listener, this makes ClearVoice an ideal everyday option, as they can transition from quiet to noisy environments without changing programs or compromising performance.

SMAKA: Can you elaborate a bit on the environments where ClearVoice is most effective?

SPAHR: ClearVoice is trying to make a distinction between speech and noise.  So any time you have a noise background that doesn’t behave like speech, ClearVoice is going to be very effective.  If you think about road noise in your car, airplane noise, fan noise, and other steady-state noises, they don’t behave like speech.  The algorithm is very effective in identifying and attenuating these types of noise.  It’s also effective when the background noise is made up of many different speech signals. In a busy restaurant for example, with a lot of people talking and a lot of reverberation, speech almost becomes a roar in the background. The algorithm does a great job in that situation. It can help you in social situations like parties, and people have reported improvements in crowded settings such as churches and auditoriums, as well.    

SMAKA: What control does the clinician have when setting ClearVoice?

SPAHR: There’s an option for low, medium and high settings.  The medium setting is probably the most popular setting, but there’s not a one-size fits all solution for how people want to hear noise.  The settings determine the maximum degree of noise attenuation implemented by the algorithm. On a low setting in the noisiest room, the noise will be attenuated by no more than 6 dB; for the medium setting, that same noise would be attenuated by up to 12 dB; for the high setting that same noise would be attenuated by up to18 dB.

SMAKA: With early noise reduction algorithms in hearing aids there was sometimes a complaint of “pumping”.  I don’t hear this so much today, but would you ever get something like that with ClearVoice?

SPAHR: No.  You won’t hear pumping with ClearVoice.  You will, however, hear a change in the noise level a couple of seconds after you transition into a noisy room from a quiet room, for example.  You will not hear pumping, but you should hear the room getting quieter.

SMAKA: How does ClearVoice affect music perception?

SPAHR: I was involved in a very early trial with ClearVoice when I was at Arizona State University. We tested it in the lab and were very confident in it, and then we let people take it home to use it in real world environments.  People came back very enthusiastic about their experience.  They expressed benefit in many expected situations and they also reported a few unexpected benefits. 

They said it helped with listening to the television, which was not surprising since a lot of television shows have background noise.  They also said it helped with the radio, which we thought might be an effect of the attenuation of car noise. What was surprising, however, was that a lot of people reported that ClearVoice improved their listening experience with music.  When you think about music, especially music with multiple instruments playing at once, from an algorithm’s perspective, it should look very similar to noise. So you might make the assumption that if music were to be attenuated, it may not be as pleasant for the listener. In this case, what seemed to be happening was that ClearVoice was pulling out the lyrics from the instrumentals, resulting in a better overall listening experience.

SMAKA: I have heard a lot of people with cochlear implants say that they don’t enjoy listening to music, oftentimes because they can’t understand the lyrics, so this sounds like an exciting development.

SPAHR: Yes, it’s very exciting.  We’ve had recipients say that they now find themselves driving their car and singing along to the radio, and they’ve never been able to do that before ClearVoice.

SMAKA:  Let’s get in to the clinical trial.  Where was it conducted?

SPAHR:  The clinical study investigated the benefits of ClearVoice in 46 adults who were experienced users of HiRes Fidelity 120 sound processing.  The participating sites were House Ear Clinic, Los Angeles, Johns Hopkins University, Mayo Clinic, Medical Universtiy of South Carolina, Medical College of Wisconsin, Midwest Ear Institute, University of Kentucky, and Washington University.

SMAKA:  What were the key findings, and where can professionals who are interested read the details?

SPAHR: Results showed that ClearVoice significantly improved speech understanding in speech-spectrum noise and in multi-talker babble without compromising speech understanding in quiet. A majority — 93% — preferred to use ClearVoice for everyday listening. Most recipients — 98% — benefitted from at least one ClearVoice setting and all recipients indicated that they liked and would use ClearVoice, which is impressive.

AB has a white paper if you want to read further details.

SMAKA: Tony, what can we expect to see in the future from Advanced Bionics, based on some of the research you’re working on? 

SPAHR: Our mission at Advanced Bionics is to improve the lives of individuals with significant hearing loss through state-of-the-art technology.  With that in mind, this is a pretty exciting time here.  As you know, a few years back we were acquired by Sonova, the same company that owns Phonak, which is one of the largest and most respected hearing instrument manufacturers in the world.  That shared ownership has given us access to a host of new technologies.  What you can expect to see is our cochlear implant system incorporating some of the proven and innovative tools that are currently used in the Phonak system.

SMAKA:  Thanks for discussing ClearVoice today.  Wishing you all the best in your new position, and hoping we can catch up again as some of these new research projects come to fruition. 

SPAHR:  That would be great, thank you.

For more information, visit www.advancedbionics.com or the Advanced Bionics Expo Page on AudiologyOnline.