Does software for video voice recognition exist?

[mazinz]

Hopefully I have this in the right forum..

I was just thinking about this today. I know voice recognition into text software exist for hands free operation, but does such software exist that you can input a video stream from a film and have it make a text file in which you can then create a subtitles file for?

In my case I have some region 2 German dvds that were actually Phillipine/Mexican action films. The dvds are dubbed in German with no english subs (I also did look to for subs online with no luck). I was thinking that it would be nice if software existed so that it would listen for the speech spoken in the film and create some sort of text or word doc from it. I can then take that, translate it into englsih (more software) and eventually make english subtitles for the film.

I suppose I could create some sort of loop where the voice recognition software would really be hearing the film soundtrack by holding the mic by the speakers, etc but figured an app for what I want to do might exist already

[gll99]

Wouldn't that be a great program if it was available?

Unfortunately the reality is that the technology is not there yet. Similar problems with speech recognition which would happen with movies also are discussed in this article

http://www.pcworld.com/article/138262/from_speech_to_text.html

[minidv2dvd]

maybe, but voice rec software normally has to be "trained" for each separate voice. might be kind of difficult. i'm not aware of any universal app, especially if it has to deal with foreign languages.

[mazinz]

Interesting article. Ehh I figured what I was after was not out there, yet anyway-

thank you both for the replies

[jagabo]

Computer scientists that have been working on voice recognition have been saying effective voice recognition is only a few years away for the last 50 years. Currently with voice recognition you have two basic choices:

1) The software can recognize a large vocabulary of a single person's speech after training. It must have clear speech without background noise. This means a head mounted microphone.

2) The software can recognize a wide variety of speakers but only for a very limited vocabulary. Think "Yes/No" or "One/Two/Three..." It will often screw up on foreign accents, American vs British English, etc.

Generalized recognition of speech in movies is very far off.

[Charmer]

Delete the post.

[filmboss80]

What about YouTube Audio Transcription? It doesn't work great, but at least it creates subtitle lines whenever a sound is heard. Is there any program like that?

I read a tech doc explaining that Google's automatic speech recognition (ASR) technology, used in combination with YouTube's caption system, "only works for English and it's been enabled for a small number of channels that usually feature talks and interviews...."

Furthermore--and this is the kicker:

"All you need to do is create a simple text file with all the words in the video and we'll use Google's ASR technology to figure out when the words are spoken and create captions for your video."

Hopefully, that will settle the matter for another 2 years (the age of this thread).

[ahhaa]

Another kludge might be to use a site like busuu.com, and find native speakers who'll collaborate...