Knowles Launched AISonic for Mobile, Ear and IoT Devices
June 24, 2019
First processor specifically designed for advanced audio and machine learning applications, enhancing power-efficient intelligence and privacy at the network?s edge.
Knowles Corporation released its IA8201, which provides voice activation and multi-microphone audio processing optimized for power-sensitive applications.
The IA8201’s host processor uses very little power, extending battery life. It also offers more accurate field voice understanding and processes commands more efficiently in environments with a lot of background noise.
Features of the IA8201 include:
- Numerous interface options and dual 175MHz cores with voltage scaling in a compact 2.6x3mm eWLB package and a 6.0x6.0mm QFN package
- Has a high compute 128-bit core (DMX) with Knowles' proprietary instruction set and a Tensilica HiFi3 core (HMD); the DMX is a 4-way floating-point SIMD processor geared toward better computing (e.g. beam-forming, barge-in, AEC), while the HMD focuses on efficient, low-power, wake-on-voice applications with a two-way floating-point SIMD processor
- Provides 1.4MB of memory for various keyword models, engines or other algorithms
- Interface options are available in 2x I2C, 2x SPI, 2 x UART, and 24 GPIOs with interrupts to support different sensor and data connections, plus 4x PDM in, 2xPDM out, 3x I2S/TDM (4ch in/out) to support up to four microphones and various audio connections
“As more of these devices get closer to the edge, the cloud’s getting pretty expensive and privacy issues are increasing,” said Mike Maia, VP of ear and IoT at Knowles. “We call ourselves the edge of the edge, right near the sensor itself, and attempt to do as much of that as possible. Keyword spotting is all done at the edge.”
Learn more at knowles.com.