About:

Intelligent-System is OpenSource hobby project. Its aim being to use modern machine learning techniques to make the world a beter place to live. It will be a signal generator replacement for small experiments at home.This perticular development till now has signal generator functionality using personal computer,and is also voice controlled.If you train and build your own acoustic model,then it also supports acoustic model switching with small modification in source code.If you would like to extend any functionality feel free to inbox me at maheshmhegade@gmail.com.

Build the project using following Sequence of commands in order

Navigate to the diretory where you have downloaded/extracted intelligent-system and issue following commands in order.

mkdir build

cp mylm.lm build/

cp mydict.dic build/

cp haarcascade_frontalface_alt.xml build/

cd build

cmake ..

make

./pcro

Functionality:

As soon as ui opens up,select detect face,it will detect face after a while(5-10 seconds) if not repeat the process untill correct face is detected.Give some name in edit box provided below and select save.

select recognize face,now system try to recognize you,and after a while system starts recognizing the voice,select parameters to generate waveform in the order wave-type,wave freqency,wave voltage,wave duration one by one by observing the terminal.

An example of voice command sequence look similar to this>>>> 'sine' > 'next' > 'one' > 'zero' > 'zero' > 'zero' > 'next' > 'three' > 'next' > 'two' > 'next' > 'generate'

In case voice recognized wrongly you can optionally say 'cancel',inorder to repeat and continue to selecting parameters.

After 'generate' utterance you should see waveform plotted displayed on ui and also hear sound corresponding to waveform,optionally you can connect headphone jack to any circuit/cro to see the waveform generated while sound is playing.

You can altogether discard voice and face recognition functionality and use as signal generator alone by mannually setting parameters and selecting Generate pushbutton.

Make-sure

You have alsasoundlib(asoundlib),opencv,sphinx-base,sphinx-training,pocketsphinx Qt installed in your system And you have mylm.lm,mydict.dic files in the build directory,where you have pcro binary file or you can have other language model if you want.

Core Idea was taken from my best friend sukruthhk@gmail.com,who is good motivator and strategist designer,programmer and architect afaik,Thanks to Him.And my project partners alok.dk.801@gmail.com and brniranjanhegde@gmail.com

Similar version of software is available for android devices in the following location. Just download install the .apk file available in the location https://github.com/maheshmhegade/PlaySound on your Android device, face/voice recognition features are still not available for android devices.