Sign in
Categories
Migration Mapping Assistant Your Saved List Partners Sell in AWS Marketplace Amazon Web Services Home Help

DeePhi Descartes Efficient Speech Recognition Engine

This is an end-to-end ASR (Automatic Speech Recognition) system with FPGA acceleration on AWS F1 by DeePhi. We modify the Baidu DeepSpeech2 framework (<a href="https://github.com/SeanNaren/deepspeech.pytorch" target="_blank">https://github.com/SeanNaren/deepspeech.pytorch</a>) for our solution of algorithm, software and hardware co-design, using LibriSpeech 1000h dataset (<a href="http://www.openslr.org/12/" target="_blank">http://www.openslr.org/12/</a>)... See more

Customer Reviews

1
  • 1
  • 4 star
    0
  • 3 star
    0
  • 2 star
    0
  • 1 star
    0
Create Your Own Review

Excellent

  • By Andrew
  • on 07/26/2018

Awesome to see the results of FPGA acceleration applied. Works flawlessly at the time of writing, and has clear documentation.

Should note that the version of aws-fpga being used on this AMI is outdated, and be easily fixed by cloning from the official aws-fpga repo and running bash sdk_setup.sh if needed.


showing 1 - 1