Update Voice Control for Banshee


I tested few speech recognition frameworks that are available free to use Simon , GNOME Voice Control and two of them had some difficulties to be used in with Banshee C# code.
GNOME Voice Control was developed for GSoc 2007 project and it seems not much work recently done on it.
CMUSphinx group of speech recognition systems developed at Carnegie Mellon University.
CMUSphinx toolkit is a leading speech recognition toolkit with various tools used to build speech applications. CMU Sphinx toolkit has a number of packages for different tasks and applications. It’s sometimes confusing what to choose.

Pocketsphinx — lightweight recognizer library written in C.
Sphinxbase — support library required by Pocketsphinx


I'm going to use these two library files to be integrated to Banshee code base.

In Ubuntu 11.10 environment :
  • To build and install SphinxBase. I downloaded directly from the repository,
% ./autogen.sh
  • then compiled and installed:
% ./configure
% make
% make install
  •   Exported environment variables:
export LD_LIBRARY_PATH=/usr/local/lib
export PKG_CONFIG_PATH=/usr/local/lib/pkgconfig
  •   Changed to pocketsphinx folder and performed the same steps
% ./configure
% make
% make install
  • Ran pocketsphinx_continuous and checked that it recognizes words.

    I tested some words and it is able find out little words. But I'm sure I need to implement a language and dictionary for it to be more accurate.

    Now My target was to integrate it to Banshee code base.

    It will be rather difficult and time consuming to build a C code to communicate to pocketsphinx. In Ubuntu system it is much easier to add GStreamer to do most work for me.  So I found a tutorial in CMUSphinx that was built to Vala port / Python describing how should we connect both GStremer pipe line and and PocketSphinx elements.

    So I read more about GStreamer. It is rather complicated. I'm on the way to develop a C Code to connect to PocketSphinx through a GStreamer pipeline and I'm going to invoke it from C# code.




Comments

  1. If all you need to do is connect GStreamer elements appropriately, you should be able to do that using GStreamer# entirely from C# without touching any C code.

    There's an experimentalish GST# backend for Banshee in trunk.

    ReplyDelete
    Replies
    1. I saw the GST# backend in Banshee. But I've already implemented come C code with help of Michael Hutchinson , my mentor Alexander K. and invoked it from C#. Now I'm developing a language model , dictionary for pocketsphinx and control Banshee play , stop functions and hopefully I expect to make a little demo video with these basic functions.
      Thanks for the suggestion.

      Delete

Post a Comment

Popular posts from this blog

GSoc 2012 Final Report

Progress of the Voice Control For Banshee