xCORE VocalFusion™ Speaker: Circular Array

Part #: XK-VF3100-C43
Silicon on Board: XVF3100-TQ128-C

The VocalFusion™ Speaker Evaluation Kit will be available on general release in October 2017. In the meantime, developers can register for early access to our Beta program at: www.xmos.com/xcorevocalfusion.

 

The xCORE VocalFusion™ Speaker kit provides direct interfacing to four PDM (Pulse Density Modulation) microphones in a circular array, with a choice of USB or I2S interfaces to connect to application or host processors. Based on the XVF3000 series voice processor, the kit provides the ideal evaluation platform for applications such as smart TVs and smart speakers.

Voice sources are isolated from unwanted noise with the integration of advance DSP techniques including beamforming, echo cancellation and noise suppression.

Keyword trigger detection delivered by Sensory enables users to activate devices simply by speaking instead of having to physically touch the device.

A rich set of optimisation parameters are available to ensure that the best results are achieved for the individual acoustics of the end product. These parameters include adjustment to noise attenuation and gain control as well as numerous optimisations for echo cancellation.

VocalFusion SPEAKER GENERAL FEATURES

  • xCORE 3100
    • High performance smart microphone for voice interfaces
      • Integrated microphone and voice DSP
      • Integrated keyword detection
      • Integrated USB 2.0 PHY for high and full-speed host and device operation
      • 2048KB on-chip flash
    • 128-pin TQFP package 0.4mm pitch
  • Microphone interface
    • Direct interfacing to 4 PDM microphones kit uses mics 1/3/4/6 only, 75x43mm rectangle
    • 16kHz sample rate
  • Audio output
    • I2S output to DAC,
    • 16kHz or 48kHz PCM
  • Host processor interface options
    • High speed USB2.0 compliant device
  • Optional I2S interface
    • With I2C for control

VocalFusion SPEAKER PRE-PROCESSING DSP

  • Beamformer
    • Adaptive de-reverberation beamformer
      • Tracks the loudest voice
      • Adaption time 100ms (typically)
    • 360° coverage
    • Up to 5m operating distance
  • Acoustic Echo Canceller
    • Mono AEC with barge-in support
    • Up to 50dB suppression
    • Initial convergence <2s
    • Adaption time 100ms (typically)
  • Noise Suppression
    • Up to 15dB stationary noise suppression
    • Up to 15dB diffuse noise suppression
  • Far-end processing
    • Dynamic Range Control
    • Equalizer

VocalFusion SPEAKER KEYWORD DETECTION

  • Sensory TrulyHandsfree™ technology
    • Evaluation license
    • Speaker independent keyword trigger detection
    • Custom triggers and multi-language support available from Sensory