Hi Karen, I've just been looking into this myself, and I am leaning toward a Sony digital voice recorder (no tapes) that comes bundled with Dragon Naturally Speaking 7 software. http://www.sonystyle.com/is-bin/INTERSHOP.enfinity/eCS/Store/en/-/USD/SY_Bro wseCatalog-Start;sid=DzPGlgoJV5zGvDW3LS_MnUUH7ntuf7kq13E=?CategoryName=pr_p_ dragon7_pa A monster URL, I know. Sorry. My understanding is that the software transcribes the recording, so that you won't have to "re-speak" the recordings of other people's voices. I haven't purchased a system yet, so I cannot vouch for its efficiency or accuracy. Best Wishes, Cassandra Van Buren ^^^^^^^^^^^^^^^^^^^^^^^^^^ Cassandra Van Buren, Ph.D. Department of Communication University of Utah 255 S. Central Campus Drive LNCO 2400 Salt Lake City, UT 84112 801.581.7268 cvb@utah.edu http://www.utah.edu/newmedia/cvb

If you are doing telephone interviews there is some software called Retail 957 which plugs into the computer (no tapes) and transcribes the interview. I'm not sure how well this works or exactly how this works as I have yet to purchase it...but it might be useful. Monica Dr. Monica Whitty Queen's University Belfast School of Psychology David Keir Building Northern Ireland BT7 1NN Phone: +44 28 90335654 Email: m.whitty@qub.ac.uk http://www.psych.qub.ac.uk/staff/whitty.html

Karen Lunsford wrote:

Hello AoIR members,

I'm hoping to hear that ethnographic transcription tools have advanced miraculously since I last was transcribing tapes. Does anyone know whether it is now possible to have voice recognition (or other) software transcribe interview tapes directly? For example, if I have a digital or analog recording, is there anything out there that can provide even a rough transcription of it automatically? perhaps like closed captioning?

NO. Contrary to what has been suggested so far, there is no voice recognition software (as its often called) that can transcribe interviews directly with any degree of accuracy that would satisfy a researcher. Even Dragon Naturally Speaking 7, which is the best out there, requires that you speak in a carefully modulated and paced way into the microphone or you'll get garbage. Also, while the most expensive version of DNS 7 will allow you to use it with more than one voice, I am fairly sure that it will only do so one-at-a-time--i.e., it won't switch itself back and forth between speaker files every time there is speaker change.

In the past, I've used Dragon Naturally Speaking to help me transcribe analog audio cassette tapes. To do this, I listened to the tape using a regular foot- pedal transcription machine and then repeated everything that I heard into the microphone attached to the computer. That way, it didn't matter that I had multiple voices on the tape and that I didn't have each speaker train the software to recognize his/her voice (not a possibility). Dragon was trained to recognize my voice, and my re-voicing of everything could produce acceptable first passes of the transcript. Then, I could go back to edit. Saved some time, but more important, it saved some of the repetitive motions of typing.

Does anyone know of a better technique now? And/or are there tools other than voice recognition available?

I don't think so. But you might also want to ask this on the forum linked to and archived at http://www.voicerecognition.com Less biased responses can also be had on the qual-software list. Be sure to check out their archive, as questions like this are asked frequently. To do this, go to http://www.jiscmail.ac.uk/lists/qual-software.html Best, Christian Nelson