There’s no getting away from voice-activated products. They’re all around us. I’m not saying that’s a bad thing, but it’s something we need to accept and embrace. Now that we agree on that, we need to agree on the path to get to those voice-enabled products.
One vendor that helping to guide developers is XMOS, with its XVF3000 family of voice processors, which provide far-field voice capture using arrays of MEMS microphones. One version of the processor, the 3010x, adds support for Sensory’s TrulyHandsfree voice-trigger technology.
If that path is still unclear for you, have no fear. XMOS is providing a development kit that provides a quick way to start developing far-field voice capture applications. The VocalFusion speaker dev kit (XK‑VF3100‑C43) includes an XVF3000 processor card and a 4-mic circular microphone array.
All XVF3000 devices come with speech enhancement algorithms that include an adaptive beamformer. That’s how they can take advantage of signals from four microphones to track a talker as he moves about the room. It also adds high-performance, full-duplex, acoustic echo cancellation.
XVF3000 devices can be integrated with an applications processor or host PC through either USB for data and control or a combination of I2S and I2C. Custom voice and audio processing can also be added using XMOS’ free development tools.