To be precise I am about to put online my second episode.
I am almost done but, after listening to an mp3 export with my phone (instead of my desktop), I find that I have been heavy handed with the editing.
I have removed numbers of ‘hummm’, sigh and other hesitations from our conversation.
I feel like those edits are noticeable and that it sounds a bit too much like sentences are a collection of words taken from different conversations (changes of tone etc.).
Not sure if my explanation above is clear.
Maybe it is a bit obvious because, for some reason, it feels like we speak faster than usual (I am considering slowing/pitching back sections) or maybe it is because I know myself those editing are there and someone else wouldn’t notice them so much.
Would anyone have something to recommend to improve the file a bit ? Something to improve the transitions between words maybe?
Next time I will be less heavy handed with my editing but I don’t feel like starting again from scratch on this one
My recommendation would be “publish and be damned”. Treat it as a learning experience. Trying to improve an over-cooked dinner is unlikely to be a good use of time - better to spend that time on making the next one better.
You are. Each time you cut out an Ummm or Urrr, the sentence gets shorter. And yes it’s exactly like patching together miscellaneous words from the last two weeks trying to make it sound natural. The rhythm, cadence and breath control of sentences is burned into the gaps between words. Mess with it at your peril. It’s one of the big things that makes telephone automated voices sound funny.
Welcome to the first AudioBook. This is where you get to the end of the first reading, realize how much better you got over the last six weeks, how sucky the first part was, and read it all over again.
It’s a very newbie thing to wear holes in the magnifier/zoom controls constantly editing every word. You may get away with that if you need to do that once in a half-hour podcast, but not multiple times a sentence. That will sound like a hatchet job. Because it is.
Instead of publishing, use this as the class on how not to read or present. Stop Umming. Theatrical “Hollywood” voices are not just recognizable for their pitch and quality, they can also read. There’s no shortage of posters demanding we tell them how to “Make Me Sound Like XXX performer.” Sorry. There’s no “Acting” setting.
This is also where ad-lib podcasts fall apart. With no general script to follow, there’s no direction or show concept if you have to do it again.
If you’re going to do non-repeatable, ad-lib podcasts, leave the errors in. Concentrate on content. There are podcasts where the performers have natural comedic timing. They could read the phone book and make it sound funny.
You know what your third podcast is going to be about right? Not the second one, the third one. How much time did it take you to produce this one?
I don’t get what you mean with your first audio book.
Your post is remarkably similar to people on their first audiobook. They get to the end and realize how much better they got over time and experience and want us to improve, through software, the first half of the book to match the second. Probably not. The successful people read it again, effectively making that their second book.
What did you do to improve the presentation? I know this seems like a Help Desk, but it really is a forum/sharing.
You dug yourself a hole with the editing. You can’t just put silent snippets back in between words to restore graceful timing because the sudden blanks will be noticed. You can put Room Tone (normal background sound) in and hope it matches. People have done that. By that time you might be further ahead recutting the original recording.
Please note we are not intentionally delaying your posts. The forum makes us authorize new posts for a while.
I will try adding a “Room tone” as you suggest before finally exporting it.
It’s one of the odd problems with not using a studio. When the studio people stop talking, the sound channel becomes dead quiet. When the home recordist stops talking, they get fan noises, echoes, street sounds, etc. etc. Collectively “Room Tone.” If you’re careful, the presentation can sound much better by adding Room Tone seconds rather than dead silent seconds. This is an ACX AudioBook recommendation. I didn’t make that up.
It still isn’t easy, but your possibility of success goes up.
And just because of the odd way you said that… I’m suggesting adding tiny room tone snippets instead of the silent snippets between words. Not to the whole show.
You know the joke everything sounds better in French? “I broke my leg and my car is on fire.” sounds wonderful in French.
“That sounds great. I’d like two of those, please.”
My only objection to the show (so far) is the echo. When you introduced Episode Two, I thought you were applying echo special effects to the presentation. You’re not are you? The room really does sound like that.
I liked the bit where you interrupt gameplay by giving each other English lessons. That was unexpected comedy.
Of course that only works if you don’t speak French. The other language joke is to translate into French and then back, but don’t translate the actual meaning. Just flip the words and get one wrong here and there. Long ago, two comedians did a short comedy act that way.
A common curse is sounding like you record in a kitchen or bathroom…or a barn.
This was a company who did a promotional sound presentation in a studio (which worked out very well) and either ran out of money or time. They decided to record a second presentation at home. How hard could it be?
It’s not easy. The pros consider modern rooms with bare floors and walls very difficult recording environments.
Somebody in Los Angeles did a successful broadcast from his garage. The joke was that you were going to visit John’s garage to do his show. It’s not a joke. It turns out having boxes and piles of junk in the room is great soundproofing. My garage also has a four-way peaked roof which breaks up echoes.
That also works better if you live on a quiet street.
We recently moved so the place is rather sober (few furnitures, not light fixtures, no frames on the walls etc.).
The second bedroom’s floor is carpet so it should help a bit.
Also we will be in a couch instead of around a table (guest knocking the table has been repeatingly an issue).
In the meantime, the third episode is pretty much 100% recorded in public spaces (a bar and a market in Barcelona).
Myself and my guest were both wearing Lavalier with the XY (I use a Zoom H4N) still recording as well.
I have listened to the rush only once (on the flight back) but I was surprised how good it seemed to sound (at least there is no echo) and the three separate tracks should allow me to lower the background sound.
Fun fact: I am an architect and I did had a full year of Acoustics forced on me. You are right about cardboxes. Actually even human bodies (guests, not corpses) do improve things a bit too. That’s why concert halls often have more reverb when there is no audience.
Exactly correct. As long as the voice is clear, putting your audience in the scene with gentle background sounds (I will not call them noise) is very effective (assuming they don’t throw you out of the restaurant). Lavalier microphones are a news gathering interview trick. You can do an interview with one long microphone and keep moving it back and forth, but that can be a nightmare later in post production. Lavaliers are much better if you have time to put them on.
I was surprised how good it seemed to sound
How good it does sound. As long as the background sound doesn’t kill you, that can be very effective. The extra, unseen performer in radio dramas is the sound effects operator.
“Rumble, rumble. Tinkle glasses, tinkle glasses. ‘May I take your order?’”
I did had a full year of Acoustics forced on me.
“No, no. Don’t force me to learn about stuff that will be valuable later.”
There is an exotic sound room design with little soundproofing, but non-parallel walls. I recorded many sound tracks in a room like that. It looked like a normal room unless you were paying attention. “The ceiling doesn’t look straight.”
When it’s not over the top (catalan coffee machine in episode 3), I actually find it quite pleasant to have a reasonable amount of background noise.
It’s totally pleasant. There are Hollywood awards given to people who can produce that effect on cue.
There are demonstrations where you get to watch a famous movie before they put all that background stuff in. It is completely flat, colorless and uninteresting.
There was an early morning broadcast radio show shot that way. I’m sure it was a nightmare to engineer the sound, but it had a great following because it sounded exactly like you were there with friends having breakfast.
This is exactly what it sounds like when someone sets up a video camcorder across the room, presses the little red record button and does a show. Echo city.
You have strike two in addition to that. The accents of the actors would normally give spice and variety to the presentation and be fun to listen to (see: everything sounds better in French), with the echo it just makes everybody hard to understand.
You need to find a “studio.”
There was a recent broadcast radio show built around a voice performer getting stuck in a closet while she was recording an audiobook.
She was overdue to perform the book, so she took her recorder and blankets and quilts into a hotel room closet, closed the door and started to record her book. Apparently, she does this all the time. She has a stand-alone recorder and makes a studio as she goes.
In this case the door locked and she couldn’t get out. The broadcast radio show was built around her efforts to get out of the closet. She left her recorder running as she was calling for help, so the show was cut between her live broadcast dialog and the closet dialog.
But she had a studio.
Quick note on Episode 3:
If you have to record while they’re grinding espresso, Just stop talking and pick up the whole sentence again after they stop. If it’s random enough cafe background noise, nobody will catch the edit. Warn the actors you are going to do that. Same if you meet someone you know and have to stop the show to say hello.
It’s going to be very difficult to beat Episode 3.
If I had to find something wrong:
— I would have panned the American interview center even if I had to do that segment in mono rather than stereo. Suddenly having the whole show coming out of one side is very odd, particularly if the listener is on headphones or earbuds. If you were completely obsessive, shoot some restaurant noise in stereo with no dialog and lay some of that in behind the mono interview.
— Grinding espresso…
— There is a trick with interstitial music. If the track has a slow, quiet build-up and then the main theme hits, start the music early so the main theme hits one half-beat after the person stops talking. Theme-under starts at 46:04, the performer stops talking at 46:10, but the main part of the song doesn’t get rolling until 46:19. There’s no statement that this is a separation, or it’s in the wrong place.
NPR Marketplace has a segment called “Daybook.” The woman who does that brief segment (or her producer) is a grand master at editing. Not only do they get the production clean, but more often than not, the music has a connection to the last story she presented. I don’t know how they do that because it’s a time-critical show.