We provide voice recognition training for people who want to get the best out of Naturally Speaking. I have not tried what you are suggesting but I would expect the error rate to be unacceptable with NS version 8 or earlier because they rely on you training the machine to recognise your voice by reading some standard text (one of the options is an extract from a Dilbert book, which is great). As someone else has said, good accuracy relies on clear, consistent speech, and low background noise is essential.
Nuance claim that version 9 does not need to be trained, and that it 'just works'. I am sceptical but have not used it so I can't say for certain. This version may be good enough to do the automatic transcription but I still suspect it won't. Steve -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Rebecca Cox Sent: 22 May 2007 11:02 To: wsg@webstandardsgroup.org Subject: [WSG] Anyone had success using Dragon Naturally Speaking to transcribe audio files? Hi there, I'm having a go at using Dragon Naturally Speaking (http://www.nuance.com/naturallyspeaking/preferred/) to automatically transcribe audio files (mp3 and wav). It's pretty patchy so far, and I wondered if anyone has had much success with this? Cheers :) Rebecca ******************************************************************* List Guidelines: http://webstandardsgroup.org/mail/guidelines.cfm Unsubscribe: http://webstandardsgroup.org/join/unsubscribe.cfm Help: [EMAIL PROTECTED] ******************************************************************* ******************************************************************* List Guidelines: http://webstandardsgroup.org/mail/guidelines.cfm Unsubscribe: http://webstandardsgroup.org/join/unsubscribe.cfm Help: [EMAIL PROTECTED] *******************************************************************