Advice on software to use for turning speech on video into a transcription

Hello,

I am wondering what programs people use to convert speech in video into a transcript. I film a lot of lectures and events that go for at least an hour. Was wondering which programs can handle large and long duration files. Am in the process of trying Premier Pro but it stalled for over 12 hours before failing to complete the speech analysis task. I use final cut for everything else.

thank you

Wanda

Joining this thread in case there is such a beast; I could use this today.

Speech analysis is not going to be instant. And what kind of footage and machine are you using? I have used it on a batch of footage with a script. That worked fairly well because I had a script to start from.

There may be a couple of solutions by Nuance, but they will be quite pricey and speech analysis produces rather high error rates.



www.strypesinpost.com

Here you go, MacSpeech Scribe:

[www.nuance.com]



www.strypesinpost.com

Or you could upload it to Youtube in a restricted area and go with the pretty instant translation they provide. It's not deadly accurate, but it hits most of the time, and you can manually adjust anything that's not right. Of course, downloading it is another thing. Not impossible though.

I use Voicebase.com. They also have a free machine transcript which isn't great. But the sight also indexes words, and can search for a word or phrase.

So I use it for my documentary work. If they need real transcripts they can pay for them, but this way i have a some sort of tool.

Thanks everyone :-)

It's the Golden Fleece among documentary tools. I've tried Nuance Scribe (actually pretty cheap, and you get what you pay for) and short of getting Ray Kurzweil to code something for us, there's always sending it to India!

[datalystllc.com]


- Loren

Today's FCP 7 keytip:
To reposition a Marker press Command and drag it!

Your Final Cut Studio KeyGuide™ Power Pack
with FCP7 KeyGuide --
now available at KeyGuide Central.
www.neotrondesign.com

There was a guy on the old AVID-L (Hulfish?) who had a method that consisted of training one of the speech recog programs to his voice then he would play the tapes in headphones and repeat aloud everything that was said. It's 1:1 timewise once you get used to it but that might be somewhere on your fast/cheap/good selection matrix.

ak
Sleeplings, AWAKE!