Training CMU Sphinx Speech Recognition Software on Ubuntu 14.04
Jason Jason Photo Jason Jason, Feb 09, 2016

CMU Sphinx is an open source speech recognition software that can run on a server. I used CMU Sphinx's adaptation tutorial as a reference for this post. Here's a quick video of me going through the process outlined below: Prerequisites: Ubuntu 14.04 CMU Sphinx installed and ready to go WAV file samples with corresponding transcriptions. Example: wav, transcription sample.fileids file with each line representing each of your sample file names. In this case (below), I named each sample wav file 1.wav, 2.wav, 3.wav, etc. and then put the filenames in the list: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 1...

Continue reading this post
Setting Up CMU Sphinx (PocketSphinx) on Ubuntu 14.04
Jason Jason Photo Jason Jason, Feb 09, 2016

PocketSphinx is CMU’s fastest speech recognition system. It’s a library written in pure C which is optimal for development of your C applications as well as for development of language bindings. At real time speed it’s the most accurate engine, and therefore it is a good choice for live applications. It's good for desktop applications, command and control and dictation where fast response and low resource consumption are the goals. Also it includes support for embedded devices with fixed-point ariphmetics and is successfully used on IPhone, Nokia Maemon devices and on Windows Mobile. You can f...

Continue reading this post