STM32 Alexa Voice Service - STMicroelectronics

Overview

Reference designs

HW Evaluation Tools

Resources

Complete and highly integrated reference designs simplifying the addition of Voice user interfaces to your embedded devices.

STM32 Voice UI includes state of the ART local and cloud-based solutions supporting far-field and noisy environments.

The Amazon-qualified, AVS for AWS IoT reference design allows rapid integration of Alexa voice services and enhances users experience with Natural Language Understanding and cloud services capabilities.

The local voice reference designs integrate Sensory’s advanced Technologies to offer most intuitive and customizable off-line device interactions.

Discover our new STM32 Audio/Voice solutions

STM32 Alexa Voice Services Solution

Amazon decided to maintain AVS for IoT system for existing products only. This solution can’t be used to develop a new product. Please contact Amazon for more information on AVS services.

STM32 Alexa Voice Services (AVS) solution eases development of simple and cost-effective natural-language understanding and voice-based user interaction for smart-home devices and smart appliances. The Amazon-qualified reference design takes smart living to the next level transforming the way users can interact with embedded devices.

Smart-home and smart appliances devices can now boost users experiences by adding state-of-the-art natural voice interaction capability and expand even more their features through smart cloud based services.

The Alexa Voice Service design kit includes Audio Front End (AFE), local Amazon "Alexa". Wake word detection and AVS Integration for IoT connectivity and achieves Amazon's far-field qualification with 36 mm microphone spacing.

Unlike digital signal processors (DSPs) and flashless processors based designs, the STM32 AVS design kit integrates necessary system features including powerful audio processing, communication interfaces RAM and Flash memories, in a single STM32H7 high-performance Microcontroller chip. This integration enables the board to have small dimensions and a simple two-layer construction for cost-effective deployment in customers’ end products.

The STM32 Alexa Voice Service solution is based on the small form factor and modular STM32 Voice UI evaluation board

STEVAL-Voice-UI offering possibility to adapt audio components selections and microphones spacing to your specific needs.

STM32 Local Voice UI solution

STM32 high Microcontrollers and Microprocessors enable state of the art, far field and noisy environment resistant, local voice command.

STM32H723 product line allows single ship support of advanced keyword spotting with Audio Front-end, wake word and up to 30 local commands recognition with no external memory. The STM32H723 embedded flash allows fast access to the commands models avoiding cache misses and enabling faster response times.

STM32MP1 achieves higher-end off-line voice UI by enabling more natural speaker commands with richer vocabularies with nested key words. It also offers the possibility to enhance even more the user interface with advanced Graphics.

HW Evaluation Tools

STM32 Alexa Voice Services Solution

STEVAL-VOICE-UI Voice User Interface reference design
- The STEVAL-VOICE-UI board is a qualified hardware reference design enabling easy and cost effective addition of the Alexa Voice Service (AVS) Integration for AWS IoT core to your smart embedded devices.
  
  The STEVAL-VOICE-UI enables you to augment your user experience with state-of-ART, hands-free, Natural Language Understanding based voice control capability addition to your device.
  
  You can talk to Amazon Alexa on your IOT end product. With Alexa, you can ask to control your smart home devices, get assistance, hear the news, check weather, play music and more.
  
  The board comes with qualified software design kit implementing all needed Audio Front End, Amazon "Alexa" Wake Word, audio playback and connectivity to AWS IoT.
  
  The software SDK runs on internal memories only offering maximum integration and allowing minimum BOM cost solution with simple PCB design
  
  The STEVAL-VOICE-UI was built with a modular approach allowing easy prototyping and debug as well as easy adaptation to your specific microphones spacing, User interface and audio output requirements.
  
  Features
  - Main board
    
    STM32H743VIT6E high-performance MCU handling AFE, local wake word, TCP/IP, Wireless and AVS for AWS IoT connectivity stack as well as audio playback
    
    WIFI subsystem including Murata 1DX module used in bypass mode and ISSI IS25LP016D 2Mbytes NOR flash hosting WIFI low level software
    
    36x65 mm, simple PCB layout
  - Microphones and speaker daughter board
    
    MP23DB01HP Microphone Mems in 36 and 30 mm spacing configurations
    
    FDA903D class D audio amplifier
    
    Leds and user buttons
    
    8 ohm Speaker

Part number	Description
STEVAL-VOICE-UI	Voice User Interface evaluation kit

STM32 Local Voice UI solution

The STM32H735G-DK Discovery kit is a complete evaluation and development platform the STM32H735IGK6U microcontroller, embedding a Cortex-M7 running at 550 MHz, 1 Mbyte of Flash memory and 564 Kbytes of SRAM. The board come with 1 microphone mems and offers different extension possibilities including a mems microphone connectors. Different microphone extension shields are available on demand and allow to evaluate the audio front end software that brings resistance to far field and noisy conditions. the board also include an audio codec and a display for more advanced user interactions.

FEATURES

STM32H735IGK6U microcontroller
ST-MEMS digital microphone
Audio mems daughter boards expansion connector
SAI audio codec
4.3’’ TFT 480x272 display
USB and Ethernet connectors

Resources

00 Files selected for download

PRODUCT SPECIFICATIONS

Description	Version	Size	Action
DB4346 Voice user interface evaluation kit	1.0	2397 KB	PDF
UM2800 Getting started with the STEVAL-VOICE-UI voice user interface evaluation kit	1.0	9.74 MB	PDF