In other words, you need to rip an audio file (wma. wmv, mp3, etc.) from the video to use these kind of software. The recognition rate of conversations would be (probably rather) lower than usual voice input method through a microphone, though.
It's so bad that when you turn of the sound and just read the subtitles, you think the person speaking is either drunk or on some very interesting candies.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.