Kaldi acknowledged as most popular framework for speech recognition
At the recent GPU Technology Conference, held in San Jose, California, NVIDIA founder and CEO Jensen Huang stated that Kaldi had become “the most popular framework for speech recognition”. And was “now optimized for GPUs”.
Kaldi, a toolkit for speech recognition, was created in 2009 at a Johns Hopkins University workshop titled “Low Development Cost, High Quality Speech Recognition for New Languages and Domains”. The success of Kaldi has lead industry hardware manufacturers to optimize it as a selling point to their consumers.