Knowles Corp. has launched a Raspberry Pi development kit that will allow engineers to bring voice, audio edge processing and machine learning capabilities to internet of things (IoT) applications.
The Knowles AISonic IA8201 kit allows designers, makers and engineers to have a single tool to streamline design, develop and test the technology.
The kit is powered by Knowles auto edge processors Open DSP that combines two Tensilica-based, audio-centric digital signal processor (DSP) cores. One core for high-power computing and artificial intelligence/machine learning (AI/ML) applications and another for low-power, always-on processing of sensor inputs.
The IA8201 has 1 MB of RAM on-chip and allows for high bandwidth processing of always-on contextually aware ML use-cases and memory to support multiple algorithms.
“According to Q3 2021 Parks Research survey data, over half of U.S. broadband households find it appealing to use voice control for their connected devices,” said Mark Vena, senior director of smart home research at Parks Associates. “The demand and the opportunity for voice applications is substantial, but it has often been incredibly complex and expensive for OEMs/ODMs to add the latest voice capabilities to devices. Solutions like those being offered by Knowles that enable voice recognition in a cost-effective, reliable, and easy to use manner will strongly resonate with consumers.”
The kit also includes:
- Onboard audio algorithms and AI/ML libraries.
- Noise elimination algorithms for partners such as Amazon Alexa, sensory, Retune and Alango.
- TensorFlow Lite-Micro SDK for fast prototyping and AI/ML product development.
- Options for two or three microphones and two microphone array boards.
- Built-in microphone array for audio and voice capabilities.