TalkTo for ST: Voice Communication

With the ST Voice Wearable reference design, DSP Concepts enables near-field voice activation on wearable devices such as smart watches, fitness bands using power and MIPS optimized 2-mic TalkTo.

Benefits

  • High quality, full duplex voice communication
    • Robust to double talk, acoustic echo and distortion
  • Cost-effective BOM for small form-factor products
    • Runs on single low-cost STM32H7 micro-controller 
    • No external memory required
  • Flexible tuning options
    • 1-mic Voice calling with advanced tuning options to trade-off between AEC and speech quality
  • Reduce tuning and certification risks and costs
    • Embedded Audio Weaver Core and Audio Weaver Designer interface with drag & drop programming environment 
    • Real-time debugging tools for expedited hardware validation

Audio Front-end Features

  • 50+ dB Acoustic Echo Cancellation (AEC) together with Echo Residual Suppression (ERS) for excellent full duplex voice quality
  • Single Channel Noise Reduction (SCNR) eliminates stationary noises such as fan noise
  • Automatic Gain Control (AGC) for optimal hearing at the far end
  • Receive Equalization and Volume control for speaker processing at the near end

Software Integrations

  • Audio Weaver Designer: Drag-and-drop graphical tool to customize and tune audio processing models in real-time with 500+ modules
  • Audio Weaver Core for ST Voice: Embedded audio processing engine for Cortex-M7
  • Voice Communication : Noise reduction, echo cancellation and automatic gain control for far-end (Tx direction) and Receive EQ and volume control for near-end talker processing

Hardware Specifications

  • Processor
    • STM32H747XIH6 Dual Core MCU with Arm® Cortex®-M7 running at 480MHz and Arm® Cortex®-M4 running at 240MHz
  • Memory
    • 2048kB Flash
    • 1024kB RAM
  • Audio I/O
    • SAI audio codec
  • Board connectors
    • Camera ( 8-bit)
    • USB with Micro-AB
    • Ethernet RJ45
    • SPDIF RCA input and output
    • Stereo headset jack including analog microphone input
    • audio jack for external speakers
    • microSD card
    • TAG-connect 10-pin footprint
    • Arm Cortex 10-pin 1.27 mm-pitch debug connector over STDC14 footprint
  • Board expansion options
    • Audio daughter board 
    • Arduino™ Uno V3,
    • Pmod™ Type 2A and Type 4A
    • STMod+

Included with the demo kit:

  • Audio Weaver Designer: Drag-and-drop graphical tool to customize and tune audio processing models in real-time with 500+ modules
  • Audio Weaver Core for ST Voice: Embedded audio processing engine for Cortex-M7
  • Voice Communication : Noise reduction, echo cancellation and automatic gain control for far-end (Tx direction) and Receive EQ and volume control for near-end talker processing

Ready to get started?

Schedule time to speak with one of our audio experts.