OT but I need help in cutting long interviews

Foregive me for posting here but it seems the most active section.

And I know it is OT.

Can anyone point me to a URL where a journalist explains how to interview someone for about an hour and extract the ten minutes of useful facts they come out with?

This must be something that people do daily when they are ghosting celebrity biographies. Maybe the more experienced journalists are more brutal than I have been at stopping the interviewee rambling and bringing him/her back to the subject.

So far what I have done is to replay the interview and use ‘marks’ to pick out the interesting bits. But that itself has taken an hour and now I have to try to work around those marks erasing the dross.

If it was text I could skim and delete - but with audio that takes far more time. I can see each one hour interview taking a day to edit and that is (a) dead boring and (b) not a good way to earn a crust.

I’m hoping someone here might either be a journalist or be able to point me to something I can read.

Sorry to be OT but I honestly don’t know where else to ask.


I am not a journalist, but how are you recording the interviews? It seems to me you should place some markers while you are recording to flag the vital pieces. Many portable recorders will do that, although Audacity cannot read the markers they save in the file.


Good point.

I’m using a Tascam DR-05. It can take ‘marks’ but I have been placing the device half way between myself and the sitter. So it is not easy to use the ‘mark’ key.

What I thought I could do is to set the stopwatch on my phone and make notes of the timings of the interesting bits alongside my list of questions. I could then use transfer those to Audacity.

But you are on the right track already.


It’s nice to think you can place markers at the important bits and have half the editing done, but that only sheers off the junk, and it only works if the interview doesn’t get insanely interesting near the end. The end of the show becomes a forest of markers.

You still have to make a show from what’s left.

You do go into the interview with question notes, right? So you can keep a clipboard and note the general importance of each question and how it went. If you’re at all on top of this, you won’t need the second-by-second markers or labels. Your notes and scanning forward through the work is usually enough to tell you where you are. Of course, if you’re doing all this six months later…

The transom.org people have terrific training videos and instructions how do to this podcasting thing. They may have some ideas.

I need help in cutting long interviews

— or —
I need to know how to cut sound for theatrical presentation. I’m pretty sure there’s still no shortcut.


You double posted.

Note how confusing that makes the thread.


Let us know how the interview goes. In both threads.


“You double posted.”

Is this aimed at me?

I marked my question OT as it is not a question about Audacity.

It is a question aimed at people, probably journalists, who have recorded conversations and want to trim them down to useable soundbites.

I realise it will be a splendid coincidence if there is such a journalist lurking around this forum - hence my marker OT.

But maybe someone knows someone…

In the other thread I will continue to ask questions about Audacity. But not in this one.


I can see each one hour interview taking a day to edit and that is (a) dead boring and (b) not a good way to earn a crust.

You can speed-up Audacity playback a bit, and there are “transcription” applications that allow you to speed-up playback without increasing pitch. (Audacity can do that too, but not in real-time so it’s not really an editing tool.)

I think most journalists take word-for-word notes as well as using a recorder . I don’t know how they do that… I don’t think they don’t teach shorthand anymore… I can’t write (or type) that fast, but it’s dangerous to rely on technology unless you have a backup device recording in parallel.

Of course, most interviews are less than an hour. And if it’s for “regular” TV or radio news, you’ll only be using a few seconds so there’s not much editing once you find the quote(s) you’re looking for.

A 1-hour program, it could easily take a day, depending on how “produced” it is. It takes an hour just to listen-through before editing and another hour to listen-through after editing, and usually at least another hour for editing. I’m sure they are spending more than a day editing a TV show like 60 Minutes.

I think you’re giving us way too much credit for thread purity.

No, you’re right. People who post with their Signal-Spritzer recorder problem while used in the left hand on Thursday with the idea of gaining thoughts from all the other Signal Spritzer users are probably out of luck.


We’re not bashful about moving postings around if you decided to post a Windows question in the Linux forum.

One of the forum elves is going to deal with your post. A mystery SS user is not going to ride out of the mist and engage in meaningful dialog. It has been known to happen that a forum elf has as a Signal Spritzer. That does happen.

Since there are a limited number of elves (sprayed over nine time zones). We’re all going to see your posts.

I’m a member of several forum families.

Forums are based on programming and developing communities where everybody knows what they’re doing and everybody is more or less active all the time; the concept did not translate well to consumer software. The software communities keep stridently insisting it works just fine because it’s cheap, inexpensive and doesn’t cost anything. No troublesome Help Desk.

Sorry. Doesn’t work like that. In the consumer world, somebody posts for fifteen seconds, gets an answer and we never see them again. In some very rare instances, a poster can stay around to be an elf. Neither of those is part of the “forum” metaphor.

Another reason you know you’re not in a real forum? The turnaround time is not two weeks.


Thanks DVD.

I am coming to the conclusion that some notes on paper are needed.

I have a list of about a dozen questions I am asking these folk (hence the hour of each session). The next time I’ll mark my copy of the list with the time each question gets asked and any keywords of unexpected responses.

I’m splitting my recording into separate segments for each question. The answers normally are to the point and then tail off into a ramble which I can junk. I have heard television interviewers say that the trick is to be silent when the interviewee replies and hope that the silence encourages them to spill a few beans to fill the silence. Maybe that works on a radio/tv interview but it does not happen often with my stuff so far.

The objective is to get the hour’s chat down to something a good typist can transcribe into text - and my gut reaction is that it needs to be condensed into less than a quarter of an hour.

With my results split into these sections I’ll then re-record the questions that define them and slot them in - as suggested on another nearby thread.