The Science of SonicCloud

SonicCloud’s Breakthrough Solution

At SonicCloud, we’ve worked with world class experts in digital signal processing, auditory neuroscience, and hearing loss to develop a breakthrough solution that automatically enhances the way you hear.

🏆   SonicCloud has been praised by top audiologists 🏆

Dr Robert Sweetow, PhD, Professor Emeritus UCSF School of Medicine, Renowned Audiologist, Creator of LACE Aural Rehabilitation program:

SonicCloud’s Hearing Assessment is better than any other internet hearing test out there. I was shocked at how accurate the audiometric configuration is.

I hear better on phone calls with SonicCloud than with any hearing aids I’ve ever worn. 
 
With SonicCloud Streaming I was hearing background vocals in a Beatles song that I’ve never heard before. And I’m a huge Beatles fan.

SonicCloud is a transformative product that allows individuals to make changes for themselves that they cannot get with any other product on the market.”

Dr. Jody Winzelberg, AuD, former Chief of Audiology at Stanford Children’s Health, and former Director of Audiology at Stanford Medical Center:

“SonicCloud is a game changer. It provides unparalleled user controls capable of enhancing sound for people with mild to severe hearing loss as well as those without hearing loss who can benefit from enhanced audibility in difficult listening situations. The hearing fingerprint assessment is user friendly and engaging, providing customizable quality to fit the listener’s unique needs. If you have difficulty hearing on your smartphone, try the SonicCloud solution.”

Jane Baxter, Board Certified Doctor of Audiology at Pacific Hearing Center:

“SonicCloud is the perfect solution for many people with hearing loss who work hard to understand conversations on their cell phones. Our patients tried SonicCloud in beta tests and reported that it was much easier to communicate and voices were more natural than with their hearing aids alone. The process of obtaining the hearing fingerprint was fun and engaging. SonicCloud is also useful for people with minimal hearing loss who need to communicate via their phones in noisy environments.”

What is the underlying science that makes SonicCloud’s algorithm so revolutionary?

Digital Signal Processing (DSP):

SonicCloud’s DSP algorithms are designed especially to target speech understanding and clarity. They are arranged in a unique structure, leveraging an unprecedented range of parameters for better hearing comprehension.

In fact, our sound processing is held to a higher standard of fidelity than what is typical for mobile applications (phone apps) and embeddable firmware (hearing aids), more closely resembling the quality found in professional music production!

Unlike music, however, real-time hearing enhancement is a far more demanding problem. Why? Because hearing involves psychoacoustic effects that unimpaired listeners simply never encounter (for example, overlapping effects of “comb filtering”; sensitivity to loud sounds at certain frequencies).

As a result, this added complexity requires special engineering attention from the ground up; rather than “off-the-shelf” solutions, such as personal sound amplification products. We’ve found that corner-cutting strategies are detrimental to the user experience.

Our DSP engineering decisions have always been informed by dedicated, results-oriented experimentation without a reliance on preconceived methods. Furthermore, due to our innovative paradigm, we have can deploy our DSP algorithms on virtually any platform — from iPhone to Laptop to over-the-web Audio — with minimal changes while maintaining a high level of performance.

Domain Knowledge:

Deep domain knowledge accrued over much time and exploration has allowed us to “rethink audiology methodology” from the ground up.

Our co-founder has a deeply personal understanding of the problem:

Larry Guterman, one of SonicCloud’s co-founders, has struggled with severe hearing loss for the past 20 years. His first-hand experience led to penetrating, long-term self-analysis and optimizations that would be difficult for someone without hearing loss to accomplish — let alone understand.

Larry’s rare, unmatched motivation to solve this problem, coupled with his physics background from his time at MIT and Harvard, has helped drive a major breakthrough for the hearing loss population.

Our team consists of world class experts in DSP and hearing loss:

With the help of leading university researchers, world class DSP engineers, and a maverick board of audiology expert advisors, we have developed a truly innovative filter bank design that cannot be easily reproduced. 

We’ve tested our technology with focus groups all across the country:

Since SonicCloud was founded, we’ve logged hundreds of hours in focus groups with subjects encompassing virtually all types of hearing loss.

Any time a subject has a technical or an audiological issue, we set up an expert tuning session with that person to address his or her specific hearing needs. Then, we apply what we’ve learned to any other subjects who may have a related issue. The net result is that we are always improving our product.

The results speak for themselves:

You can clearly see the efficacy of our technology as demonstrated in surveys, testimonials, and standardized word discrimination tests that we’ve conducted.

In one focus group 88% of participants reported not just “comparable” but “better” hearing — both in terms of clarity and naturalistic fidelity — with SonicCloud than with their current method of talking on the phone (including premium, Bluetooth-connected hearing aids)

Revolutionary Audiology Methodology:

We consolidate the entire audiology workflow — visiting an audiologist, taking a mold of your ear, manufacturing a hearing aid, visiting an audiologist for a second time, getting your hearing aids fitted, testing your hearing aids in a live setting, and conducting further ongoing visits to better tune your hearing aids — into a single 10 to 15 minute process that is accurate, flexible, powerful, enjoyable and effective.

This process covers 3 levels of optimization: Mechanical, Neural and Situational Tuning. Using this groundbreaking tiered methodology and the attendant proprietary technology, the user can converge on a precisely-tuned hearing profile without any audiological expertise whatsoever.

Think of it as a pyramid with 3 tiers of tuning…

Tier 1: Your Baseline Hearing Assessment (Mechanical Tuning / Data Capture)

The baseline hearing assessment tests your minimum threshold and loudness tolerances at each frequency.

The results are 5 times more granular than a typical hearing assessment- people with moderately severe to profound hearing loss can detect a difference of 2 dB, sometimes even down to 1dB increments. 

This methodology not only contributes to better speech understanding, but also mitigates “phantom” audio effects. 

With your accurately captured “mechanical” data, we can then “thread the needle” of your dynamic range to sculpt a hearing response for speech discrimination that is unmatched. How? By doing the following:

  • Deploying compression threshold values that are more precisely tied to loudness discomfort and thus audible dynamic range.
  • Optimizing time-domain parameters (typically not controlled by audiologists) which exert further impact on speech discrimination.
  • Employing a unique and proprietary clipping mitigation strategy, based on deep domain knowledge, that allows us to push the limits of enhancement beyond other solutions without leading to undesirable processing artifacts.
  • Applying a novel windowing strategy that allows us to stretch the perceived limits of the digital dynamic range of the smartphone.
  • Empowering users to adjust these parameters themselves — in the comfort of their home — via our highly intuitive and accurate user interface.

Once the baseline hearing assessment is complete, the user will have created a default hearing profile that will sound great in the vast majority of situations.

Most hearing assessments stop here. We go two steps further — Neural and Situational Adjustments— to ensure the very best speech understanding for people with hearing loss:

Tier 2: Advanced Refinement (Neural Optimizations)

Two people who present with the exact same audiogram mechanically may nonetheless have very different speech understanding capabilities, based on for example, how long they’ve had the hearing loss and how well their brains have adapted to it. Optimizing for brain differences can be a crucial factor for achieving good speech understanding. 

Our patented, innovative, interactive optimization accounts for the neural differences by giving the user, who is the only one who can truly perceive what’s going inside his/her brain, the ability to iteratively hone in on the perfect presentation level for key phonemes, by providing audible context of those phonemes. 

Here’s how it works: A phrase or word is played back on the phone while you optimize interactively — adjusting letters or phonemes important for understanding by stretching or shrinking them until they sound perfectly clear. Working through a range of representative phonemes, you can cover the key parts of speech and voila! You’ll have an accurate profile ready for use. 

Audiologists have praised our self-tuning methodology, as it allows users to achieve sophisticated contextual tuning — using easily understood visual metaphors — in a way that is not possible with other methods.

At this point, the user will have accounted for everything but the situational component. To do so, we will need tier 3:

Tier 3: Real-Time Tuning (Situational Optimizations)

Not all voices and environments are created equal.

So while your default profile will work great in most situations, it might not be optimal for when, say, you’re talking to your niece whose voice has a very high fundamental frequency, or when your friend is in a noisy restaurant with lots of chatter. That’s where the situational tuning component comes into play.

It’s the only method currently on the market that allows users to tune to specific voice or environments. 

And the interface is intuitive — for example, if the audio sounds too tinny to you, simply tap on the ‘too tinny” mini-tutorial to guide you through how to make an appropriate correction on your “hearing curve.” The same goes for other aurally challenging situations or voices. 

Once you’ve made the necessary adjustments to account for that particular voice or environment, simply save that profile. Next time you’re talking to that same person or find your friend in that same environment, your profile is ready to use!

So, what level of improvement can you expect to hear? 

Let’s look at what our co-founder hears with and without SonicCloud as an example:

Before and After Using SonicCloud:

Here’s what our co-founder hears when he’s not using SonicCloud…

And here’s what our co-founder hears when he is using SonicCloud…
Pretty amazing! Ready to hear the incredible difference for yourself…?

Download the app and take our free hearing assessment to get started!

Get Started For Free!