The reference design, originally presented at the recent Embedded World 2022 trade show in Nuremberg, Germany, allows users to activate 39 different washing machine settings through 472 unique voice command phrasing options without the need for a cloud-based voice assistant. View the reference design in action by clicking the image below or here.
Fluent.ai is a leading name in embedded speech recognition solutions, and its partnership and collaboration with Knowles Corporation has already resulted in multiple solutions for voice applications in different product categories, including a development kit for ultra low-power true wireless earbuds. This time, the collaboration between the two companies focused on fully offline, noise robust and multilingual/multi-accent voice control for washing machines and other home appliances.
The patented speech-to-intent technology from Fluent.ai leverages natural and flexible voice commands to trigger actions in multiple classes of appliances and is a perfect complement for Knowles advanced micro-acoustic microphones, speakers, edge-AI and audio processing solutions.
"The use-case possibilities are endless with our joint solution for embedded voice control with Fluent.ai. From our original partnership integrating our technology into TWS earbuds, we have since evolved it to the control of smart home appliances. The current experience of controlling your appliances with voice using a smart speaker offers a poor customer experience – one which is difficult to install and has latency as the appliances have to wait while the voice command makes its round trip to cloud and back. Moving the audio processing to the edge embedded in white goods solves this to create an easy-to-use and responsive voice interface. Fluent’s technology helps us fit voice UI into a cost-effective module with a large library of voice commands and phrases to control white good appliances," says Mehul Kochar, Sr. Director Business Development, Audio Solutions at Knowles.
Essential to household appliances, Fluent’s software supports more commands and phrasing flexibility in a small footprint. The white goods reference design improves the washing machine user experience through hands-free, low latency offline voice command control that protects user privacy, all while providing a noise robust solution to cancel out background appliance sound and other noise. The software resides locally on Knowles’ AISonic Audio Edge Processor IA8201, which does not use data from the cloud, meaning Fluent’s technology works completely offline, without the need for a Wi-Fi connection.
This implementation provides a seamless, no lag, exceptional voice control solution for white goods appliances. The combined technology can support any language or accent, allowing OEMs to ship a single product SKU to cover a wide geographic and linguistic market, and provides significant BOM cost reduction for OEMs looking to add voice recognition to their home appliance products.
"We are excited to grow our partnership with Knowles, a global leader in audio solutions, and expand the use of our combined technology to white goods appliances," says Probal Lala, CEO of Fluent.ai. "Fluent’s software is noise robust, which is key for filtering through loud background noise of appliances such as washing machines, dishwashers, kitchen range hoods and more. Our embedded technology enables users of any language or accent to command appliance tasks hands-free and without the need for internet connection. We expect this collaboration will also significantly reduce hardware BOM costs for OEMs due to the elimination of components required for cloud-based voice recognition solutions."
www.fluent.ai
www.knowles.com
