Pocketsphinx: Speech Recognition in ROS

Pankaj (~Pankaj-Baranwal)


8

Votes

Description:

PocketSphinx is an open-source lightweight speech recognition engine developed at CMUSphinx, specifically tuned for handheld and mobile devices, though it works equally well on desktop environments. Most of its features are designed to work offline. So, you don't need to be connected to the Internet when working on speech recognition applications.

ROS or Robot Operating System is an open-source, meta-operating system for your robot. It offers vast inter-operability and modularity for writing and running software for robotic applications.

But there isn't much available in the ROS universe for speech recognition. Even the official ROS package for poketsphinx was developed back in 2011 and it's completely outdated in today's time and age.

Hence, this session will provide an introduction to a better, well maintained and much advanced alternative package of pocketsphinx in ROS under development at CMUSphinx.

Prerequisites:

  • Love for Robotics and Open Source
  • Must have Linux-based OS
  • Basic understanding of Python
  • Basic understanding of ROS workspaces would be preferable but is not a necessity.

The session would also entail a hands-on experience with speech recognition in ROS. So, if you want to have a better understanding during the session, you should have ROS pre-installed in your system.

Content URLs:

ROS Tutorials: https://wiki.ros.org
CMUSphinx Tutorials: https://cmusphinx.github.io/wiki/
Updated ROS package for pocketsphinx: https://github.com/cmusphinx/ros-pocketsphinx

Speaker Info:

Pankaj Baranwal
Fell in love with open source at a young age.
Contributor at CMUSphinx.
Developed the updated ROS package for pocketsphinx.
Spent a lot of time experimenting with ROS & Robots.

Section: Embedded Python
Type: Dev Sprint
Target Audience: Intermediate
Last Updated:

No comments added so far.

Login to add a new comment.