Interview with Michael Nilsson Ph.D, Sonic Innovations, Salt Lake City, Utah
Share:
Topic: Sonic Innovations new product - Innova
Beck: Good Morning Michael. Hard to believe, but it's been about 5 years since you and I last did an interview!
Nilsson: Hi Doug. Nice to join you again!
Beck: If you don't mind, I'd like to tell the readers a little about your professional background and then we'll talk about the new Sonic Innovations product -- the Innova.
Nilsson: Sure thing, Doug.
Beck: I know your Ph.D. was in Psychology from the Cognitive Science Department at the University of California at Irvine, and you were at the House Ear Institute from 1990 to 1998 where you helped develop the HINT (Hearing In Noise Test, Nilsson et al., 1994, J.Acoust. Soc. Aa., 95(2), 1085-1099), which is a set of recordings used to measure speech intelligibility in quiet or noise. And you've been at Sonic since 1998, since the first product launch?
Nilsson: That sounds about right. The first launch was September, 1998, and I got here a few months earlier.
Beck: I believe you're a patent holder for the noise reduction algorithm used at Sonic Innovations?
Nilsson: Yes, my engineering colleague, Xiaoling Fang Ph.D., and I patented the noise reduction technology used at Sonic.
Beck: Which product generation are you guys about to introduce?
Nilsson: The Innova is based upon our third generation DSP. The first generation was Natura, and then the second generation came out in 2001, that was the Natura II which added our noise reduction, and Innova is the third generation.
Beck: I know there are many new features, but is the Innova actually a whole new chip?
Nilsson: Yes it is. Although Innova uses our same underlying core philosophy regarding how to amplify sound for sensorineural hearing loss, there are many changes.
For example, with regard to how sound gets into the chip, the Innova has two A-D converters which allows a "Dual Omni" design and 8 combinations of two inputs. For example we can use Telecoil & Omni, Dual Omni and other options, too.
Also, rather than the previous 9 channel system, we now have a 16 channel system, allowing narrower channel design. The major advantage to 16 channels is the independent activity of the 16 channels. In other words, each channel has it's own noise reduction, feedback canceller, directionality, expansion and compression ability, so we can do more "real time" analysis and processing, simultaneously.
The combination of changes form what we are referring to as our S.M.A.R.T. technology. This refers to our core system that monitors sound, analyzes the level and SNR, etc. and responds transparently to provide users with the best signal possible. We don't want a system where you can hear it changing to try and optimize settings.
Beck: Michael, let's go back to the dual omni directionality?
Nilsson: Well, Doug, as you know, there are essentially two ways to create a directional microphone system. One can use a physical filter between mic ports to create a delay in the sound delivery between the two ports, or you can accomplish the same thing electronically. In the Innova, we do it electronically. In this way, the clinician can choose between common polar plots to maximize directionality. After that, our proprietary system called "Directionalfocus" helps compare phase differences between the two mics, estimates the location of the dominant sound in each of the 16 channels, and then the gain is manipulated to maximize the sensitivity of the system to sounds from the front, while attenuating sounds from the sides and behind.
Beck: What is the total bandwidth of the circuit itself, and what are the spectral constants of the channels?
Nilsson: In the low bands the channels are 375 Hz wide and they're 750 Hz wide in the high frequencies. The total bandwidth of the chip is 10 thousand hertz, but the mic and receiver limit the practical output to less than 8 thousand hertz.
Beck: OK, so going back, each feature acts independently in each of the 16 channels?
Nilsson: Yes, that's correct. The noise reduction, feedback control, compression, directionality and other features can be controlled within each channel, to maximize performance. I should note that the directionality features are available in all shell configurations from ITC to BTE, and VCs are available from ITC to BTE too. The circuit is available in a CIC with a programming button, but the size constraints prohibit the dual omni mic ports in the CIC.
Beck: Have you tested actual patients with this technology, and do you have outcomes data available to support the technology?
Nilsson: Yes. Excellent question. We are undergoing trials with the Innova and we hope to publish the results very soon. The preliminary results and comments have been quite good, and we're very pleased. We're still cleaning up the software and the algorithms, but it's coming along very nicely.
Beck: Do you have expansion in this circuit?
Nilsson: Yes, thanks for asking about that. Innova has "Noise Limiting" and it's based on the combination of noise reduction and expansion. Expansion is a terrific tool in the Innova, and it's an important part of the noise limiting system. Expansion is particularly important to help reduce soft sounds, such as the hum of a refrigerator or other soft annoyances. Once the input gets louder, noise reduction takes over and attenuates the channels where unmodulated sound is present. In Innova, noise reduction increases as the loudness of the noise increases, so this is adaptive, or level dependent noise reduction.
Beck: Are the vents in the Innova wider diameter than previously available?
Nilsson: Yes, the vents are wider. We're able to make the vent wider across the whole product line because we have better feedback control. Further, larger vents allow less occlusion and more natural sound quality, and better immediate patient acceptance during the first fitting.
Beck: And as far as quantitative tests in the lab?
Nilsson: The lab results are great. We have some fabulous bench-top data that confirms the output signal-to-noise ratio of the devices is actually better than the input signal-to-noise ratio by over 10 dB with the appropriate features turned on. This technique for looking at changes in signal-to-noise ratio has been with us for many years, and was originally used with the telephone sciences many decades ago.
Beck: Is the T-Coil programmable?
Nilsson: Yes it is, and the T-coil is calibrated to the input of the omni microphone, so the T-coil circuit reflects the way the hearing aid is programmed. There are 4 memories and we also have "Smart Tones" to help the patient know if they're in program one, two, three or four, and they can tell if the VC is going louder or softer, and we also have a power-on delay to avoid feedback. There are also audible "low battery" signals...so all in all, the Smart Tones really are useful and they help the patient to better understand what the hearing aid is doing!
Beck: Michael, this is really very exciting technology. Before I let you run, would you please tell me a little about how the feedback management system, works?
Nilsson: The feedback system has an inverse phase filter, which is somewhat similar to other feedback cancellers, but our feedback canceller can be "trained" without hooking it up to the software, using the push-button retraining program. So it can be retrained "on the fly" as cerumen or other environmental issues impact the sound path. You can also train the canceller in preprogrammed devices without having to hook up to the software, allowing feedback cancellation in preprogrammed devices. That reminds me, we have a new cerumen protection system too, and this will be tremendously beneficial in keeping cerumen out of the instrument. The No Wax System is our new standard, but we have other cerumen options too. The Innova will have less plugging-up of the receiver because we are moving to screen-less receivers; it'll be a more reliable instrument with less potential for failure based on cerumen issues.
Beck: Michael, thanks for the update on the Innova, this is very exciting product.
Nilsson: Thank you too Doug. Nice to visit with you again.
-----------------------
For more information, visit the Sonic Innovations website.
Nilsson: Hi Doug. Nice to join you again!
Beck: If you don't mind, I'd like to tell the readers a little about your professional background and then we'll talk about the new Sonic Innovations product -- the Innova.
Nilsson: Sure thing, Doug.
Beck: I know your Ph.D. was in Psychology from the Cognitive Science Department at the University of California at Irvine, and you were at the House Ear Institute from 1990 to 1998 where you helped develop the HINT (Hearing In Noise Test, Nilsson et al., 1994, J.Acoust. Soc. Aa., 95(2), 1085-1099), which is a set of recordings used to measure speech intelligibility in quiet or noise. And you've been at Sonic since 1998, since the first product launch?
Nilsson: That sounds about right. The first launch was September, 1998, and I got here a few months earlier.
Beck: I believe you're a patent holder for the noise reduction algorithm used at Sonic Innovations?
Nilsson: Yes, my engineering colleague, Xiaoling Fang Ph.D., and I patented the noise reduction technology used at Sonic.
Beck: Which product generation are you guys about to introduce?
Nilsson: The Innova is based upon our third generation DSP. The first generation was Natura, and then the second generation came out in 2001, that was the Natura II which added our noise reduction, and Innova is the third generation.
Beck: I know there are many new features, but is the Innova actually a whole new chip?
Nilsson: Yes it is. Although Innova uses our same underlying core philosophy regarding how to amplify sound for sensorineural hearing loss, there are many changes.
For example, with regard to how sound gets into the chip, the Innova has two A-D converters which allows a "Dual Omni" design and 8 combinations of two inputs. For example we can use Telecoil & Omni, Dual Omni and other options, too.
Also, rather than the previous 9 channel system, we now have a 16 channel system, allowing narrower channel design. The major advantage to 16 channels is the independent activity of the 16 channels. In other words, each channel has it's own noise reduction, feedback canceller, directionality, expansion and compression ability, so we can do more "real time" analysis and processing, simultaneously.
The combination of changes form what we are referring to as our S.M.A.R.T. technology. This refers to our core system that monitors sound, analyzes the level and SNR, etc. and responds transparently to provide users with the best signal possible. We don't want a system where you can hear it changing to try and optimize settings.
Beck: Michael, let's go back to the dual omni directionality?
Nilsson: Well, Doug, as you know, there are essentially two ways to create a directional microphone system. One can use a physical filter between mic ports to create a delay in the sound delivery between the two ports, or you can accomplish the same thing electronically. In the Innova, we do it electronically. In this way, the clinician can choose between common polar plots to maximize directionality. After that, our proprietary system called "Directionalfocus" helps compare phase differences between the two mics, estimates the location of the dominant sound in each of the 16 channels, and then the gain is manipulated to maximize the sensitivity of the system to sounds from the front, while attenuating sounds from the sides and behind.
Beck: What is the total bandwidth of the circuit itself, and what are the spectral constants of the channels?
Nilsson: In the low bands the channels are 375 Hz wide and they're 750 Hz wide in the high frequencies. The total bandwidth of the chip is 10 thousand hertz, but the mic and receiver limit the practical output to less than 8 thousand hertz.
Beck: OK, so going back, each feature acts independently in each of the 16 channels?
Nilsson: Yes, that's correct. The noise reduction, feedback control, compression, directionality and other features can be controlled within each channel, to maximize performance. I should note that the directionality features are available in all shell configurations from ITC to BTE, and VCs are available from ITC to BTE too. The circuit is available in a CIC with a programming button, but the size constraints prohibit the dual omni mic ports in the CIC.
Beck: Have you tested actual patients with this technology, and do you have outcomes data available to support the technology?
Nilsson: Yes. Excellent question. We are undergoing trials with the Innova and we hope to publish the results very soon. The preliminary results and comments have been quite good, and we're very pleased. We're still cleaning up the software and the algorithms, but it's coming along very nicely.
Beck: Do you have expansion in this circuit?
Nilsson: Yes, thanks for asking about that. Innova has "Noise Limiting" and it's based on the combination of noise reduction and expansion. Expansion is a terrific tool in the Innova, and it's an important part of the noise limiting system. Expansion is particularly important to help reduce soft sounds, such as the hum of a refrigerator or other soft annoyances. Once the input gets louder, noise reduction takes over and attenuates the channels where unmodulated sound is present. In Innova, noise reduction increases as the loudness of the noise increases, so this is adaptive, or level dependent noise reduction.
Beck: Are the vents in the Innova wider diameter than previously available?
Nilsson: Yes, the vents are wider. We're able to make the vent wider across the whole product line because we have better feedback control. Further, larger vents allow less occlusion and more natural sound quality, and better immediate patient acceptance during the first fitting.
Beck: And as far as quantitative tests in the lab?
Nilsson: The lab results are great. We have some fabulous bench-top data that confirms the output signal-to-noise ratio of the devices is actually better than the input signal-to-noise ratio by over 10 dB with the appropriate features turned on. This technique for looking at changes in signal-to-noise ratio has been with us for many years, and was originally used with the telephone sciences many decades ago.
Beck: Is the T-Coil programmable?
Nilsson: Yes it is, and the T-coil is calibrated to the input of the omni microphone, so the T-coil circuit reflects the way the hearing aid is programmed. There are 4 memories and we also have "Smart Tones" to help the patient know if they're in program one, two, three or four, and they can tell if the VC is going louder or softer, and we also have a power-on delay to avoid feedback. There are also audible "low battery" signals...so all in all, the Smart Tones really are useful and they help the patient to better understand what the hearing aid is doing!
Beck: Michael, this is really very exciting technology. Before I let you run, would you please tell me a little about how the feedback management system, works?
Nilsson: The feedback system has an inverse phase filter, which is somewhat similar to other feedback cancellers, but our feedback canceller can be "trained" without hooking it up to the software, using the push-button retraining program. So it can be retrained "on the fly" as cerumen or other environmental issues impact the sound path. You can also train the canceller in preprogrammed devices without having to hook up to the software, allowing feedback cancellation in preprogrammed devices. That reminds me, we have a new cerumen protection system too, and this will be tremendously beneficial in keeping cerumen out of the instrument. The No Wax System is our new standard, but we have other cerumen options too. The Innova will have less plugging-up of the receiver because we are moving to screen-less receivers; it'll be a more reliable instrument with less potential for failure based on cerumen issues.
Beck: Michael, thanks for the update on the Innova, this is very exciting product.
Nilsson: Thank you too Doug. Nice to visit with you again.
-----------------------
For more information, visit the Sonic Innovations website.
