# Qualcomm's AI audio chip enables smartphone-free intelligent earbuds New chip allows wearables to run AI assistants and apps independently with enhanced audio features - **Source:** [Latest from Live Science](https://www.livescience.com/feeds.xml) - **Published:** 2026-09-25 - **Topic:** [Data & AI](https://brifing.si/c/data-ai) - **Original article:** - **On Brifing:** https://brifing.si/s/6934ff75-0974-4c72-b88a-24fb46aad646 ## Standalone AI processing The Snapdragon Sound Elite Gen 2 chip enables earbuds and other wearables to host apps directly on device memory, eliminating the need for constant smartphone connections. This breakthrough allows AI assistants to perform tasks like calendar management and online purchases without phone tethering. The chip's architecture supports two-way conversations with virtual agents and can sync with cloud services like ChatGPT when Wi-Fi is available. Qualcomm demonstrated this capability at their annual Snapdragon Summit in Maui. ## Expanded wearable applications Beyond traditional earbuds, Qualcomm plans to integrate the chip into various 'hearables' including over-ear headphones, audio glasses, and camera-equipped buds. The technology will also power future AI wearables like pendants and clips. These devices will passively capture environmental audio and video data to build contextual awareness for AI systems. Qualcomm assures privacy through low-resolution cameras and AI masking that obscures identifiable features in captured footage. ## Dual-connection technology The chip introduces micropower Wi-Fi channels alongside Bluetooth, consuming significantly less energy than standard Wi-Fi at the cost of higher latency. This dual-connection approach improves audio quality since Wi-Fi doesn't compress data like Bluetooth. A specialized subsystem enables lossless audio streaming typically reserved for premium services. The system can maintain connections to smartphones or computers through visual interfaces when needed. ## Advanced noise control Machine learning algorithms power adaptive noise cancellation that users can control vocally. Unlike traditional noise cancellation, this system lets wearers selectively hear specific sounds like emergency sirens or conversation partners while blocking other noise. The feature builds on University of Washington research that isolated individual voices in noisy environments using earbud-embedded computing systems. ## On-device intelligence An embedded neural processing unit (eNPU) enables always-on AI features including real-time transcription, translation, and contextual awareness building. The system continuously monitors the user's environment to enhance personal profiles for future queries. While audio-focused, the chip interfaces with various sensors and cameras, creating a comprehensive data harvesting system for agentic AI applications. ## Technical specifications The new chip doubles the AI processing power of its predecessor, performing 128 billion operations per second compared to 64 billion. It's 30% smaller and uses 40% less power. The onboard memory allows app storage directly on devices, with each having its own IP address for direct server communication via HTTPS. This eliminates the need for intermediary phone or computer connections for many functions. ## Industry context Qualcomm's technology competes with emerging solutions like Viaim's conversation-recording devices and xMEMS' piezoelectric audio chips. Apple may be developing similar camera-equipped AirPods for environmental data collection. The chip represents an evolution of Qualcomm's 3-year-old S7 Gen 1 system, addressing previous limitations in power consumption and app hosting capabilities for wearables. ## Implementation challenges While Qualcomm promises balanced performance between audio quality, AI features and battery life, continuous operation will impact battery drain differently than intermittent use of conventional earbuds. The company is exploring running multiple apps simultaneously in RAM. Device makers will initially control which apps are pre-bundled, creating a new ecosystem for wearable applications similar to early smartphone app stores. ## Sources covering this story - https://www.livescience.com/technology/electronics/new-kind-of-ai-audio-chip-transforms-earbuds-into-always-on-intelligent-devices-that-run-without-a-smartphone-how-does-it-work — --- _Summary written by Brifing from the original reporting. Copyright in the original article belongs to its publisher; read the full piece at the link above._