3

how to perform continuous speech recognition from microphone using pocketsphinx .how to use gstreamer plugin api in C ?

user1462684
  • 91
  • 2
  • 9

1 Answers1

4

how to perform continuous speech recognition from microphone using pocketsphinx

There are several APIs for that. One of them is plain pocketsphinx API. You can read pocketsphinx_continuous source in pocketsphinx/src/continuous.c to see how to use it. You can just include source code from continuous.c into your application.

how to use gstreamer plugin api in C ?

If you want to work with gstreamer, you first need to learn gstreamer basics. Here you can find some gstreamer documentation.

http://docs.gstreamer.com/display/GstSDK/Basic+tutorials

Unfortunately it's hard to cover everything in a single response, but the basic sequence of items is:

  1. Create pipeline with audio src, audio resampling, vader and pocketsphinx
  2. Connect to pocketsphinx plugin signals, the signals will provide you the recognized data
  3. Start the pipeline and the main loop.
  4. Wait for the recognized data signal and perform required action.

It's not different from the python example in pocketsphinx/src/gst-plugin/livedemo.py except you need to call the same methods from C.

Nikolay Shmyrev
  • 24,897
  • 5
  • 43
  • 87
  • Thanks for reply.. the problem is my application requires some library , which conflicts with continuous.c . I m using SFML library . gcc -o continuous continuous.c -DMODELDIR=\"`pkg-config --variable=modeldir pocketsphinx`\" `pkg-config --cflags --libs pocketsphinx sphinxbase` -lsfml-graphics -lsfml-system – user1462684 Feb 05 '13 at 13:34
  • What exactly conflicts there, sorry, you need to provide more information. I suggest you to use CMUSphinx forum to ask question like this. – Nikolay Shmyrev Feb 05 '13 at 20:22