Electronics Engineering Herald                 
Home | News | New Products | India Specific | Design Guide | Sourcing database | Student Section | About us | Contact us | What's New
Processor / MCU / DSP
Memory
Analog
Logic and Interface
PLD / FPGA
Power-supply and Industrial ICs
Automotive ICs
Cellphone ICs
Consumer ICs
Computer ICs
Communication ICs (Data & Analog)
RF / Microwave
Subsystems / Boards
Reference Design
Software / Development kits
Test and Measurement
Discrete
Opto
Passives
Interconnect
Sensors
Batteries
Others

New Products

  Date: 02/08/2015

Programmable logic based voice interface for IoT wearables and mobiles

QuickLogic has announced a multicore programmable logic-based SOC chip for sensor interface applications in smart, wearable and IOT devices. This chip offers hardware flexibility for sensor interface applications and also for always-on voice processing applications.

To enable always-listening voice recognition by using this SOC chip, QuickLogic has partnered with voice recognition software export Sensory, Inc.

The voice trigger is basically handled by hard system blocks, which are designed to consume less current of 350 microAmps for always-on voice trigger. The Voice commands works without the support of cloud.

The logic circuitry inside the chip accepts digital input from Pulse Density Modulation (PDM) and Inter-IC Sound (I2S) microphones to feed PDM to Pulse Code Modulation signals to the Sensory's TrulyHandsfree software for voice processing. Also hard coded is Sensory's Low Power Sound Detector (LPSD) technology, which allows the speech recognizer to be suspended while an ultra-low power sound detector is running and listening for what could be speech.

The SoC chip also supports voice recognition of noisy voice commands, which can be always-listening fixed triggers, enrolled fixed triggers, user defined triggers and passphrases, and up to 20 phrase spotted commands that can be accurately detected in silent to extremely noisy environments, claims QuickLogic. Since due to the use of customised hardware for voice processing the power consumption is far less compared to CPU based software processing of voice data.

Initial samples of the EOS platform with integrated voice processing are expected to be available in September 2015.




 
ADVT
Home | News | New Products | India Specific | Design Guide | Sourcing database | Student Section | About us | Contact us | What's New
©2010 Electronics Engineering Herald