Synaptics has developed an innovative approach to integrating large language models (LLMs) into household appliances, demonstrated through a prototype dishwasher interface. By extracting common user inquiries from a dishwasher manufacturer’s user guide, they created a system capable of understanding and responding to voice commands without requiring internet connectivity. This offline functionality ensures user privacy and reliability.
—
Synaptics is my Embedded World 2025 video coverage sponsor, check out my Synaptics videos here: https://www.youtube.com/playlist?list=PL7xXqJFxvYvhAbQoe9YN4c84SqXxIY3fQ
—
The core of this system is the Synaptics SL1680 processor, a highly integrated AI-native system-on-chip (SoC) optimized for multi-modal consumer, enterprise, and industrial IoT workloads. The SL1680 features a quad-core Arm® Cortex®-A73 64-bit CPU subsystem, a multi-TOPS neural processing unit (NPU), and a high-efficiency GPU for advanced graphics and AI acceleration. These components enable efficient on-device processing of complex AI tasks, such as natural language understanding and speech recognition. citeturn0search1
In the prototype, the system performs speech-to-text conversion to interpret user queries. It then vectorizes these queries and matches them against a pre-processed set of frequently asked questions using the NPU’s capabilities. Once the best match is identified, the system generates a voice response through text-to-speech synthesis, providing users with immediate assistance. For example, when asked, “How can I put plastic on my dishwasher?” the system responds with guidance on placing dishwasher-safe plastics on the top rack to avoid melting.
This integration of LLMs into appliances like dishwashers offers potential benefits beyond user convenience. Manufacturers could reduce the volume of calls to customer service centers by enabling appliances to address common user issues directly. Additionally, this technology could be applied to other devices, such as set-top boxes, to assist users in troubleshooting without external support.
The SL1680’s architecture supports running multiple AI models in parallel, allowing for a range of functionalities within a single device. This flexibility means that while the current implementation focuses on specific use cases, future developments could expand the system’s capabilities without significant hardware changes. For instance, users might issue commands like “Do it in a quick mode” without pressing any buttons, streamlining appliance operation.
Localization is another advantage of this technology. The system can be programmed to support multiple languages and understand various accents, making it adaptable to diverse markets. This adaptability ensures that users worldwide can interact naturally with their appliances, enhancing the overall user experience.
Updating the system with new information or responses is straightforward. Manufacturers can deploy updates over the internet or via mobile devices, ensuring that appliances remain current with the latest user guides and troubleshooting procedures. This ease of update allows for continuous improvement of the system’s responsiveness and accuracy.
The cost implications of integrating such advanced AI capabilities into appliances are manageable. Since the SL1680 SoC is designed for efficiency and performance, manufacturers can incorporate these features without substantial increases in production costs. This balance ensures that consumers receive enhanced functionality without a significant price hike.
Incorporating LLMs into household appliances represents a significant step toward more intuitive and user-friendly home environments. By leveraging Synaptics’ SL1680 processor, manufacturers can deliver smart appliances that not only perform their primary functions but also assist users through natural language interactions, setting the stage for a new era of intelligent home devices.
Check out all my Embedded World 2025 videos in this playlist: https://www.youtube.com/playlist?list=PL7xXqJFxvYvjgUpdNMBkGzEWU6YVxR8Ga
This video was filmed using the DJI Pocket 3 ($669 at https://amzn.to/4aMpKIC using the dual wireless DJI Mic 2 microphones with the DJI lapel microphone https://amzn.to/3XIj3l8 ), watch all my DJI Pocket 3 videos here https://www.youtube.com/playlist?list=PL7xXqJFxvYvhDlWIAxm_pR9dp7ArSkhKK
Join https://www.youtube.com/charbax/join for Early Access to my videos and to support my work, or you can Click the “Super Thanks” button below the video to send a highlighted comment!
