The Race towards Smarter Hearing Aids

In Part 1, we saw why a hearing aid’s AI has to run on the device itself and how much battery life it can cost. Bringing that cost down is now the main challenge. This is pushing brands to innovate and update their AI-flagship platform almost every one to three years. Phonak, Starkey and GN update yearly, while Oticon and Signia generally go two to three years between platforms.
But not every “AI hearing aid” uses AI the same way, though. In some, AI only recognises the listening environment and picks a sound setting. In others, it separates speech from noise in real time and is central to how the device works. The form factor often shows which one you’re looking at, because a dedicated AI chip needs room to sit in.

Smart Hearing Aid in Various Form Factors

Source: Counterpoint Research

The AI Architecture Race:

The global hearing-aid market is concentrated among five major players: Sonova, Demant, WS Audiology, GN Store Nord and Starkey. In August 2026, four of these companies introduced new AI-focused flagship platforms within two weeks, each taking a distinct approach to on-device intelligence.

• GN’s ReSound Sensia combines real-time environment classification, four-microphone beamforming and automatic DNN noise reduction.
• Demant’s Oticon Reveal uses two AI systems in parallel, with one focused on speech and the other on preserving and balancing surrounding sounds.
• Sonova’s Phonak EON Audio Sphere is powered by the new HYPERSONIC chip, which uses 37% less power and enables a device that is 25% smaller and 20% lighter than the previous generation.
• WS Audiology’s Signia MaX combines four task-specific neural networks on a new chip that delivers 54 time more processing power, 50% greater energy efficiency and twice the memory than its predecessor.
• Starkey took a different path with Omega AI, launched in October 2025, and recently launched Omega AI+ combining neural-network-based directional processing with generative AI assistance and health-tracking capabilities.

Top AI-Powered Hearing Aids and Their Capabilities

Source: Counterpoint Insights
* Phonak Audéo EON Sphere is 25% smaller than its previous generation, so it carries a smaller battery. Its new chip uses less power to run AI

Each company runs it AI on its own branded chip platform with proprietary algorithm trained on its own data. Each one is taking a different path like Sonova leads on efficiency, WS Audiology on raw computing power, GN on automatic environment handling, Demant on balancing speech with surrounding sound, and Starkey is betting on health features beyond hearing. However, they all face the same challenge of packing this intelligence into a device where battery life, size and processing power are tightly limited. The key question is how much useful AI can be delivered per milliwatt.

Consumer tech catching up:

Beyond traditional hearing-aid players, consumer tech brands are also entering the market. Apple already offers an FDA-authorized Hearing Aid feature on AirPods Pro, while Samsung has received FDA clearance for its Galaxy Buds Hearing Aid feature, with rollout planned for Q4 2026. Meta also launched its Ray-ban Meta Audio which features an open-ear earbuds, microphone and AI-assistant built into the glasses. Meta is also bringing a Hearing Enhancement feature to supported AI glasses. The FDA-cleared software feature is designed to help adults aged 18 and older with perceived mild-to-moderate hearing loss hear better. It will launch in the U.S. later in 2026 at $149.99, with users able to set it up at home without a clinic visit or prescription

These developments are blurring the lines between consumer electronics, wearables and hearing devices. By integrating hearing support into mass-market earbuds and glasses, these products could make basic hearing assistance more accessible while intensifying competition in the OTC hearing-aid segment. Traditional hearing-aid players are also participating, with GN offering Jabra Enhance and WS Audiology partnering with Sony on OTC hearing aids.

Outlook

With more than 500 million people projected to have disabling hearing loss by 2030, demand for smart AI hearing aids is set to rise. According to Counterpoint’s Global Edge AI Enabled Wearable Forecast, prescription hearing-aid sales are projected to exceed 31 million units by 2030, growing at a 4% CAGR from 2025. Nearly 50% of these will feature Edge AI capabilities.

This will push the hearing aid brands to focus on three core areas: efficiency, miniaturization and intelligence. More efficient AI will bring advanced processing into mid-tier devices, so features found only in today's flagships will reach more affordable price points. Better chips will allow increasingly powerful models to fit into smaller form factors, and the current top line products will be available in the mid-tier. Intelligence will also extend beyond sound, turning hearing aids into sensing platforms that combine audio processing with health monitoring and generative AI.

As dedicated AI silicon becomes increasingly central, competition will shift from adding AI features to building architectures that deliver more intelligence within strict power, size and latency constraints. AI is becoming a meaningful advancement in hearing aids because it addresses one of the technology’s hardest problems: understanding speech when the world around the wearer is noisy and unpredictable. However, the future is not about maximizing AI compute, but about optimizing it for real-world consumer value. Every additional unit of processing must justify its impact on speech understanding, listening effort and personalization against the cost in battery life, device size and price.

1st Part of the series:

From Amplification to Edge Intelligence: The Rise of AI in Hearing Aids