Paul Neil, vp of marketing, noted: “According to Gartner, there will be 500 smart devices in every home by 2025. Today’s solutions don’t scale, so there is a need for a new interface. Because you won’t be able to control these things from a smartphone or tablet, voice is the new click.”
The XVF3000 family is said to feature an updated DSP core to handle ‘heavy lifting’ for speech processing, with additional processing for echo cancellation and beam forming.
While the XVF3000 supports four microphone inputs, the XVF3101 adds keyword recognition. Neil said these devices will be the first in a line of products addressing markets ranging from speakers to industrial applications. Both variants, whic come with 512kbyte of RAM and 2Mbyte of flash, are supplied in 128pin TQFPs.
XMOS has also launched the VocalFusion Speaker development kit, which includes an XVF3000 processor board. Two variants are available: circular, intended for smart speaker applications; and linear, for smart TV and sound bar projects.
