Hi I want to have a speech recognition api or sdk which recognises the speech spoken by the user and gives it\'s text form.
Detailed Description is as follows:
You can take a look at https://github.com/KingOfBrian/VocalKit, but I have not tried it myself.
You can try
http://www.politepix.com/openears/
As for speed, it should be fast, you probably don't use it properly. As I understood you have text already and you need to build grammar from this text.
You can also try Nexiwave.com.
I think the function you are looking for is what we can TimeStamping: http://nexiwave.com/index.php/applications/for-transcription-companies
It basically take an audio and the text, we then put timestamp on each sentence and word.
Ben