Tokyo PC Users Group
	  Home Page
Home
Members Only
Newsletter
Newsgroups
Become a Member
Meeting Info & Map
Officers
Members
Corporate Members
Photos
Workshops & Training
Other Clubs
Job Hunting?
Constitution

Hour 25 Internet Radio

Reprinted by Permission submitted by David Bernat

Greetings From the Starscape.

I'm pleased to let you know that this week's edition of Hour 25 with author Dan Simmons is now online at www.hour25online.com.

Dan met with us when he was on tour for his new book A Winter Haunting. It's a freestanding horror novel, but it revisits the events in his book Summer of Night. You don't need to be familiar with Summer of Night, but if you've read it, you'll get even more out of reading A Winter Haunting. Dan said he sees the two books as a kind of Moebius loop. All the events in A Winter Haunting change every central thing in Summer of Night.

He also talks about the protagonist from his book Darwin's Blade, the real life events that are portrayed in it and the possibility of a TV series based on it.

The book he currently writing has some truly amazing characters -- a 14 ton robot on the moon Io with a passion for Proust and Shakespeare who works with another robot on the ice of Europa who has a passion for Shakespeare's sonnets. Now, there's an image!

I got distinct impression that he's developed a great love for some of the characters he's created, such as Dale Stewart and Duane McBride from A Winter Haunting and Darwin Minor from Darwin's Blade, not so much because they were his own inventions, but because they were born from the framework he created, then took on a life of their own and became fully realized human beings.

Dan writes in multiple fields -- science fiction, horror and suspense and does them all well. A former teacher, he speaks well on the subject of writing. Tune in. I think you'll agree. And if you like what you hear and you're a fledgling writer, you'll be pleased to know that Dan is one of the writers at the upcoming Clarion West. Applications to attend run until April 1st, so check out the link on the Hour 25 web page if you're interested.

For all you folks that have been asking how the show is recorded and edited, I've written you a description of our recording and editing process, below. However, if this sort of information isn't your kind of thing, feel free to stop reading and get on with your life. Hopefully, you'll tune in to hear Dan Simmons.

It's been a long day. I just finished breaking down the recording equipment and editing the closing for this week's show for the SECOND time today. It's like this: Although the Dan Simmons interview was recorded a couple of weeks ago and edited during the week, when the time came to record the show opening, we ran into a problem. Warren had developed a bad case of laryngitis and clearly wasn't going to be recording anything, so I did the opening and the closing of the show.

Now, this isn't something I do too often, and so I opened the computer file for the last time I did one (24 December 2001), and copied the text for the show closing with a few minor revisions. Unfortunately, I failed to put in one major item, and we didn't catch it until the show closing was recorded and uploaded to the test version of our web page. Then Warren listened to the closing and came out of his office and croaked out, "You forgot to thank Dan Simmons for the interview in the closing." Of course I did. The closing I used from last December had no guests. There were only readings from authors who I probably should have thanked, but, as they were long dead, I was fairly certain they weren't going to receive my thanks even if I'd offered them. So, once again, we broke out all the recording equipment and re-recorded and edited the closing.

So, for all you folks that have been curious about how the show is recorded and edited, this makes a fairly good example. First of all, why are we packing and unpacking all the recording gear? Well, when we removed the show from Pacifica radio, we had already put the show on the Internet and upgraded our recording equipment from analog to digital. But we still weren't quite sure what our recording setup was going to be. We had initially thought we would do the show much in the same way we had done it at the radio station, so we set up a recording area and left it set up. But this didn't end up being the way we recorded shows. About 90% of all the interviews ended up being done in the field, not in the studio. We would travel to the hotel rooms of guests on book tours. We'd interview people at conventions, conferences and other events. For local authors, we'd travel to their homes. The important thing was to pick anywhere that was most convenient for the authors. This suits authors on tour particularly well, because their book touring schedules have grown increasingly jam-packed. Often, they're in town for only a day or two, and it could be on any day of the week, not necessarily the day of the week we used to air the show live. So the recording equipment is now kept packed into two compact cases for location travel, and only set up as needed.

We record on digital minidiscs that hold 74 minutes of audio. I will say that one real advantage of recording in studio is that while we are recording, we are simultaneously transferring the digital file of the interview onto the hard drive of one of the networked computers where the interview will be edited. Interviews recorded in the field have to be transferred to the computer for editing in real time. That means a 90-minute interview takes 90 minutes to load into the computer. This is time consuming, but as we often don't know when an interview will air, it isn't practical to store everything recorded on our hard drives. We sometimes have as much as six months worth of shows waiting to be aired, so you can imagine how much hard drive space that would require! We are considering the possibility of simultaneously transferring the file onto the hard drive of a laptop while recording in the case of interviews that need to be aired that night. (Such as convention interviews.) Then the file can be transferred via the computer network instead of in real time.

Minidiscs are reusable, but we never reuse them. We archive the interview on disc. Before the show is over, it will ultimately be saved and archived in several versions and formats, all of which will be eventually archived on CD ROM.

The interview file is saved with a name indicating it's a raw (unedited) copy of the interview, then saved as the edited version. So in case the editor screws up the edited version, any damaged portions can be copied from the raw version and pasted into the edited version.

This is basically what I did with this week's show closing. I did completely re-record the closing segment, complete with the usual couple of words that I'd stumbled over and had to re-read for editing. But before I re-edited the NEW show closing, I tried to see if I could just edit in the missing bit. It's a lot like editing a word document, but instead of highlighting, copying and pasting in words, you're doing the same to sound waves.

There are a number of audio editing programs out on the market. The one we currently use is called Cool Edit. I opened up the new show opening file in Cool Edit, highlighted and copied the sound waves for the lines, "I'd like to thank Dan Simmons for taking the time to talk with us, and I'd like to thank the Group Mind." Then I opened the old show opening, and pasted the lines in where I wanted them and deleted the old line thanking the Group Mind. After listening to the sound waves before, during and after the new section to see how it sounded, I adjusted the volume of the new section slightly to match the audio level of the rest of the closing. Done!

Editing an interview takes a bit more finesse. Our general rule is, one interview, one editor, because it requires developing an ear for the nuances and speech patterns of the guest. Sometimes, Warren does most or all of the editing for the show, other times, I do it. Warren almost always edits the show opening and closing. Currently, I'm editing most of the interviews. This has to do with the fact that I have more time at present than he does. At other times, he's done it all. One thing we both agree on -- we both have the same obsessive attention to detail, so no matter who edits the show, we'll both be happy with the quality of the work.

Some interviews, like this week's, could have been run with virtually no editing. If I had snipped out the usual bit of nattering that starts out first thing on every interview, "Say a few words so I can check the sound level." etc., I might have been able to run the Dan Simmons interview as is. It really only required a little adjustment in a few places where the sound levels were too loud. Most interviews require much more work. The average interview will lose 10%-15% in length, but no actual content will be removed from it. (An aside about that opening nattering while the sound level is checked. Last week, when asked to say something, Howard Hendrix said something to the effect of: "Greetings to the web based listeners of Hour 25." It was great, and I thought I should clip it out and save it to do something with. However, I broke my cardinal editing rule. If you're going to do anything to any part of the interview, do it now. Once you've moved on to the next section, you're NOT going to come back later. I forgot about it entirely, and it was included and uploaded. When Warren listened to the interview to make sure all the parts played, he said, "I like that greeting to the listeners from Howard." And I said, "Yikes! I forgot about it. I was going to take it out, because it was the sound level check. It was pretty cool, though.")

Interviews are recorded in stereo, with each person on a separate audio track. So when the editor listens with headphones, they will hear one person in each ear. On the computer screen, the editor sees two sound waves, and a time line running below them. In theory, each audio track can be edited separately, but in practice, this seldom works because if the two people are speaking with any significant volume and are sitting reasonably close together, then what Warren says will be picked up at a very low level by the guest's microphone and vice versa. Once in a while, this doesn't happen, and if both people talk at the same time, the editor can separate what each person says. However, sitting people that far apart is rather antisocial, and doesn't lead to the rapport that makes for a good interview, so we don't worry too much about trying to separate what people are saying on the rare occasions when they talk over each other. I'd rather have great content than easy editing. I guess I'm more of a producer than an editor.

It's easy to say that now, but after I've been editing for 20 hours, I think differently. Twenty hours? Yes, you wouldn't believe how long it takes. My rule of thumb is to allow one hour of editing time on average for each seven minutes of finished interview. If the interview needs some major adjustment in sound levels, plus basic editing, (The past two shows have needed this.) then maybe five minutes per hour of editing. Phone interviews are often two minutes of finished interview per editing hour because the audio quality isn't good though the phone line, which is why we tend not to do them. We are researching the purchase of additional audio equipment that would improve the audio quality of phone interviews.

What am I doing in all this time? First, I just look at the entire interview as a whole. If it looks like it should, one track should have more sound waves than the other one. At the very least, the two tracks should look about the same. In other words, I should be able to look at the two tracks without listening to them and say, this one has more dialog - that should be Dan Simmons. Because a good interview is always about the guest, not the host. Then I look at the volume of the tracks. Does it have large amounts of sound waves that are off the scale and need to come down in volume? Is the reverse true? Are either or both of the audio tracks so low in volume that they need to have the volume increased? Usually you end up with both in interviews. Some bits too loud, others too soft. These will be adjusted as I look at each part of the interview. Sometimes the volume of the entire track or the entire interview can be increased or decreased at once, but that's usually the exception.

After deleting the nattering at the front of the recording, I start listening from beginning to end, in very small increments. How small? Depending on the work that needs doing, I view and listen to sections that have been enlarged to fit the screen often only 2-5 seconds at a time. An interview like this week's with little work required can often be viewed in 10-12 second segments. First, I adjust the volumes if needed. Sound waves that are off the meter (redline on the scale) are highlighted and the volume is reduced to 90%. Sections that are too soft are highlighted and amplified by whatever volume I think is appropriate to what is being said, but is more audible. This could be anywhere from 115% - 200% increase, however, you have to listen to thE Interview before and after the increase to see if the sound level change sounds abrupt or obvious or if any background noise increases so much that the volume change is apparent. Much of what I do in editing involves making changes, then listening to the section over again to see if it sounds right.

After volume changes, the thing I do most is edit, as in delete. How can I delete anything while claiming not to take out any content? Because I want listeners to get involved in what's being discussed, a number of real life distractions are edited out if they occur. The phone ringing. Doorbells. Room service. Sirens. Coughs, and Warren usually coughs at least a few times each interview. (Don't worry, his health is just fine.) The best I can do about this if the guest is talking is to drop the volume way down on Warren's mike, so that you only hear a little cough that bleeds through on the guest's mike. This is also what is done when both people talk at once. Ninety-nine percent of the time, it's obvious that only one of the things being said is important. Usually, one person breaks off without finishing and waits to have a turn to talk. So the unimportant words are dropped in volume so you can hear the important part.

Words that are taken out include, ur, um, and occasionally you know, if it occurs multiple times in a sentence. Also, if anyone stutters, and most people do sometimes (myself included) it is deleted. Unfortunately, stuttering is frequently contagious, so if the guest does it, Warren does too. If Porky Pig were a guest on Hour 25, he would say, "Th-th- th-that's all, folks!" but when his words were aired, you'd hear, "That's all folks!" People also unintentionally repeat words and sometimes, have false starts to sentences. These all end up on the cutting room floor, so to speak. But more often, I tighten up the size of the pauses or spaces between words or sentences. This is more of an art than a science. The editor must learn the cadence of the person's speech in order to preserve the subtleties of their personal expression. Often the interview is done late at night, guests are tired, and this helps pick up the energy level of the interview. Thankfully, Cool Edit has the ability to let you UNDO what you've done an almost endless number of times, so you can keep trying different ways of editing. If someone repeats a word, "But but what I really meant" does it sound smoother if you delete the first BUT, or the second one? You can try it both ways before you decide.

Done correctly, the editing process simply takes out those foibles of speech that your mind ignores when talking to a friend. These are also elements that would be deleted if the interview had been transcribed into text. Take a look at one of the wonderful interviews in Locus Magazine. You won't find Ums, or false starts to sentences. I tell guests that properly edited, the interview will sound exactly like themselves, but they'll sound smoother, less hesitant in places and a bit more erudite. As a rule, guests can't identify any of the editing, and they're pleased with the way they sound.

While I'm editing, I'm also taking careful notes. I record the starting length of the raw interview, and the time code that indicates the end of the section I've completed editing. In other words, I want to stop for a moment, so I save the interview file and note that I've edited through 34:28.022. I also make notes about the content of the conversation so I have some interesting points to mention on the web page and in the newsletter. I also make notes of the exact time when the conversation starts a new topic. Why do I do this? Well, that dates back to the days when Hour 25 was on live radio. We were taught that the average listener tuned in and out to a radio station every 15 minutes, therefore we should re-introduce the guest and topic that often. We've adapted this for our new format on the web. We certainly don't have to re-introduce the guest, because the listeners won't be tuning into the middle of the broadcast, and the guest and topic are clearly identified on the web page. We also have the luxury of producing a show that is of any length we feel covers the guest or subject in depth. We don't have to squeeze the entire show into a 58 or 116 minute window.

However, listeners may not have time to listen to a two hour interview with a guest, so we've taken that 15 minute segment from the live radio format and adapted it to the web. Hour 25 interviews are broken into parts of roughly 15 minutes each. They play continuously, but if a listener isn't able to listen to an entire interview, he or she can stop and come back later. If you remember the number of the part you've already heard, then you can jump ahead to the number of the next part. When you start listening again, part 1 will start playing. Just click on the number of the part you want to begin playing, such as part 4 and you can pick up pretty much where you left off. To try to make the show easier to follow for folks who can't listen to it in one session, I try to break the parts at the beginning of a new topic.

After the editing is completed, the two stereo tracks (one of each speaker) are converted by the computer program to a 16 bit monaural track. The mono track is broken into roughly 15 minute parts, saved and renamed. Each part is then converted into an MP3 track. The tracks are then uploaded to the server of our web host in Philadelphia.

All in all, in both my roles as the producer and as the member of the staff that does most of the editing, I have a vested interest in making sure we have interesting guests. Because while you spend an hour or two listening to the interview, I spend several days with it. I hope you feel your time with us is well spent.

Join us next week when we'll have an all science show. Our guest will be Dr. Brian Marsden, Director of the Minor Planet Center, and we'll also bring you a brief excerpt from Dr. Stephen Hawking's recent CalTech lecture. And I have to tell you, hearing Stephen Hawking in his own words is truly awe inspiring.

Until next time, take time to dream, but turn your dreams into actions. The future is what you make it.

Suzanne

February 17th, 2002

Listen to this week's edition of Hour 25 at

http://www.hour25online.com

Science Fiction Radio for Southern California since 1972

Now Available World Wide Over the Internet



© Algorithmica Japonica Copyright Notice: Copyright of material rests with the individual author. Articles may be reprinted by other user groups if the author and original publication are credited. Any other reproduction or use of material herein is prohibited without prior written permission from TPC. The mention of names of products without indication of Trademark or Registered Trademark status in no way implies that these products are not so protected by law.

Algorithmica Japonica

October , 2002

The Newsletter of the Tokyo PC Users Group

Submissions : Editor


Tokyo PC Users Group, Post Office Box 103, Shibuya-Ku, Tokyo 150-8691, JAPAN