Audéo Sphere™: Advancing speech understanding with AI

Phonak introduces the world’s first hearing aid with a dedicated artificial intelligence (AI) chip that addresses the number one client need – speech understanding in noise.

You’ve likely heard the buzz around the launch of Sphere hearing aids, but you may not have had the chance to experience them yet. The excitement centers on the enhanced speech understanding in noise, made possible by Sphere’s dedicated AI chip.

You might be wondering, ‘Is the improved speech understanding in noise really worth the hype?’ The answer is yes, and here’s why:

Speech understanding in noise is one of the most challenging listening situations for people with hearing loss and one of the most important predictors for hearing aid satisfaction.1 Hearing aid wearers require a better signal-to-noise ratio (SNR) compared to normal hearing listeners for the same speech intelligibility performance.2

Although directional microphone technology has previously addressed the issue of speech understanding in noise, its effectiveness in this area is now remarkably refined through the integration of the world’s first AI-driven chip—available in Phonak Audéo Sphere™ Infinio devices.

In this article, we will explore how Spheric Speech Clarity brings a significant breakthrough in allowing more access to speech cues in challenging listening environments.

Spheric Speech Clarity

Phonak Audéo Sphere™ Infinio is the world’s first hearing aid with a dedicated artificial intelligence (AI) chip. The DEEPSONIC™ chip hosts a deep neural network (DNN) algorithm that significantly suppresses the background noise, leading to remarkable speech intelligibility improvements and reduction in listening effort.3,5

Spheric Speech Clarity improves the SNR in any direction. Spheric Speech Clarity is combined with directional microphone technology so that your client receives the benefit from two speech enhancement technologies.

Unprecedented SNR improvement

A recent study demonstrated the progression of Phonak technologies. Figure 1 shows the SNR benefit compared to an omnidirectional microphone mode. The higher the SNR, the better the hearing aid is at separating speech from noise, resulting in clearer speech for the client.


Figure 1. Progression of Phonak technologies: Signal-to-noise ratio (SNR) improvements with respect to the omnidirectional microphone mode for Phonak technologies averaged across three realistic scenes, including a café scene, dinner party scene and food court scene. For details, see Raufer et al. (2024).4

With a fixed directional microphone setting, 2.9 dB SNR benefit can be achieved in realistic listening situations for a speaker from the front. With the introduction of noise-cancelling algorithms, like NoiseBlock and Dynamic Noise Cancellation (DNC), the SNR benefit could be improved from 2.9 dB to 5.1 dB.

StereoZoom 2.0, in combination with NoiseBlock and DNC, offers an additional SNR improvement, totaling 6.4 dB. This underlines that the well-known state-of-the-art directional microphone and noise reduction technologies from Phonak are instrumental in building success for clients who use Phonak hearing aids.

Spheric Speech Clarity goes beyond existing hearing technology. Spheric Speech Clarity, in combination with a fixed directional microphone, offers an unprecedented SNR improvement of 9 dB at default strength and 10.2 dB at maximum strength. The strength of Spheric Speech Clarity can be adjusted by the hearing care professional in Phonak Target and by the client using the myPhonak-App.

What this means for your clients

In social settings, such as crowded restaurants or family gatherings, background noise can make it difficult to follow conversations and stay engaged. With improved speech clarity in noise, your clients can participate more actively, reducing the strain associated with trying to understand speech in challenging environments.

To date, Spheric Speech Clarity is Phonak’s best performing technology allowing the client to focus on what’s relevant while having conversations in challenging listening environments. Sphere improves the signal-to-noise ratio, significantly improves speech intelligibility, and reduces the listening effort.5

To learn more about Phonak Audéo Sphere Infinio and experience it yourself, we invite you to visit the Phonak website.

References:

  1. Abrams, H. B. & Kihm, J. (2015). An introduction to MarkeTrak IX: A New Baseline for the Hearing Aid Market. Hearing Review, 22(6).
  2. Killion, M. C. (1997). The SIN report: Circuits haven’t solved the hearing-in-noise problem. Hearing Journal, 50(10), 28-32.
  3. Diehl, P.U., Singer, Y., Zilly, H., Schönfeld, U., Meyer-Rachner, P., Berry, M., Sprekeler, H., Sprengel, E., Pudszuhn, A., & Hofmann, V. M. (2023). Restoring speech intelligibility for hearing aid users with deep learning. Sci Rep; 13, 2719. https://doi.org/10.1038/s41598-023-29871-8
  4. Raufer, S., Kohlhauer, P., Uhlemayr, F., Kühnel, V., Preuss, M., Hobi, S. (2024) Spheric Speech Clarity proven to outperform key competitors for clear speech in noise. Phonak Field Study News, retrieved from https://www.phonak.com/en-us/professionals/audiology-hub/evidence-library.
  5. Wright et al. (2024). Spheric Speech Clarity applies DNN processing to significantly improve speech understanding from any direction and reduce the listening effort. Phonak Field Study News, retrieved from https://www.phonak.com/en-us/professionals/audiology-hub/evidence-library.

Co-Authors:

Stefan Raufer, Audiological Researcher at Sonova

Stefan joined Sonova in 2021 and is developing audiological concepts for new hearing aid features. Together with other teams, Stefan brings latest technology to our products, making sure that new technologies translate to user benefits. Stefan holds a PhD in speech and hearing sciences from Harvard University.

Philipp Kohlhauer, Audiological Researcher at Sonova

Philipp joined Phonak HQ as an Audiological Researcher in 2021. With his background as a hearing aid specialist, he supports the development of new features to ensure their quality and optimum performance. Philipp did his Master of Science in Auditory Technology at the University of Lübeck.

Frédérick Uhlemayr, Audio & Development Engineer at Sonova

Frédérick joined Sonova in 2021 and is currently working on perceptual models and tools for technical measurements. Frédérick graduated as Diplom-Tonmeister from the University of Music and Performing Arts in Vienna. Frédérick has a background in professional audio reproduction as well as in automotive audio.

Volker Kuehnel, Principle Expert Hearing Performance at Sonova

Volker Kuehnel, PhD, holds a doctorate in physics and completed his studies in 1995. From 1995 to 1997 he worked in Oldenburg as a research assistant in the Medical Physics group of Prof. Dr. Dr. B. Kollmeier. Since 1998, he has been working at Phonak/Sonova in product development on the audiological design at the interface between hearing aid algorithms and fitting software. His work focuses on the audiological quality of hearing systems to achieve maximum customer benefit.

Michael Preuss, Audiology Manager at Phonak HQ

Michael joined Phonak HQ as an Audiology Manager in 2020. Benefiting from his lecturing background at the Academy of Hearing Acoustics in Lübeck and living with hearing loss, Michael delivers comprehensive expert training sessions and provides audiological input during product development. Michael studied Hearing Acoustics at the University of Applied Sciences in Lübeck, Germany.

Shin-Shin Hobi, Senior Product Manager for Audiological Performance at Phonak HQ

Shin-Shin joined Phonak HQ in 2006 and has worked on various projects as an Audiology manager. In her current role as Senior Product manager for Audiological Performance, she ensures end user and hearing care professional needs are taken into account during product development of audiological features. Originally from Australia, Shin-Shin earned her Audiology qualifications from University of Melbourne. She gained wide clinical experience in private practice, in Perth, before making the move to Switzerland.