I know that "x-webkit-speech" is able to do some sort of speech recognition, and actually the recognition is pretty good.
I find that "x-webkit-speech" is good for developing a Q&A application, where you just ask question and the system recognises the question and answer you.
However, what i am looking for is a way to perform continuous speech recognition in the browser. For example, if I am hearing a lecture online, I would like to transcribe what the professor is talking about in real-time.
Is it possible with "x-webkit-speech"? My feeling is that "x-webkit-speech" automatically stops when it detects a short silence period, which is annoying. I know it actually converts the audio into flac and send it back to google server. Is it possible to send back a longer audio? I need continuous speech recognition on the web.
