Editing What First? Second?

Post your questions & get help from friendly LibriVoxers
Post Reply
KinoOnik
Posts: 34
Joined: July 5th, 2019, 9:53 am

Post by KinoOnik »

Hello all.

I have recorded a couple of copies of a single reading I was given the "OK" on. [English Literature for Boys and Girls, Chapter 38 "The Death of Sir Thomas More"]

I know there is "editable" (?), content. Long lapses (silence), and the invariable flub(s), etc.

Noise Reduction question... Will "Noise Reduction" eliminate long pauses? Wait a sec., I don't think so. If it did, it would eliminate the standard 5 sec. of silence req'd at the end.
If I'm correct, "Nevermind." (Emily Litella, portrayed by Gilda Radner of SNL Fame [RIP]).

What should deal with with first? Second?, etc.

As of this writing, I only know how to snip silences. And I think I can do noise reduction.

If, while recording I make an obvious "flub" and realize it straightaway, can I keep the "Camera running". Take a nice pause I am certain to see, then pick it up from before the flub, Take another long pause? During editing, I can erase the "flub" then trim the silence off the ends of the redo? I'll call this "Edit on the Run". (Assuming of course it is thought a good idea).
TriciaG
LibriVox Admin Team
Posts: 60576
Joined: June 15th, 2008, 10:30 pm
Location: Toronto, ON (but Minnesotan to age 32)

Post by TriciaG »

Correct - Noise reduction does not truncate silences.

When I have a list of corrections to make after submission and PL, I start at the end of the recording and make the edits back to front. That way the time stamps given me do not change when I add/remove stuff.

Otherwise, there's not really any order to making edits after it's been PL'd.
If, while recording I make an obvious "flub" and realize it straightaway, can I keep the "Camera running". Take a nice pause I am certain to see, then pick it up from before the flub, Take another long pause? During editing, I can erase the "flub" then trim the silence off the ends of the redo? I'll call this "Edit on the Run". (Assuming of course it is thought a good idea).
Yep. Or clap (or have an air horn :P) to make a spike in your recording that you'll be sure to see (even more visible if you have "show clipping" checked - I think it's in the View menu).

I am a slow-poke editor. I read, make all my flubs as I go, and correct them by repeating the phrase (sometime with a pause, sometimes with very little pause) while I record. Then (here's the slow part) listen to the whole thing and cut out all the bad parts.

There's also a cool technique called Punch and Roll where you stop the recording, put your cursor at the start of where the mistake happened, hit a key combination... and Audacity will delete from the cursor (deleting the mistake), then back up and play the last few seconds of recording, starting to record where the cursor is. I personally cannot do it, but others find great success with it. Here's a quick video about how one reader does Punch and Roll: https://www.youtube.com/watch?v=K1_Gti3lHaM
Serial novel: The Wandering Jew
Medieval England meets Civil War Americans: Centuries Apart
Humor: My Lady Nicotine
KevinS
Posts: 15477
Joined: April 7th, 2019, 8:32 am
Contact:

Post by KevinS »

You didn't ask specifically about truncating silences, but I'll share something anyway.

I use truncate all the time. If you 'select all' but then pull the 'curtain' back from the end so as to unselect your 5 seconds of silence. You can truncate at will. I usually truncate anything equal to or over 1.5 seconds down to 1 second. It seems to work for me. If I have more than 1.5 seconds silence in the rest of my recording, it's not planned. Either the cat jumped in my lap or something else distracted me.

Projects are different, though. If I'm reading a series of poems, for instance, I'll sometimes leave 3 seconds of silence between each piece. More than 3 seconds usually sounds like an eternity to me, though I've worked on projects where it's been appropriate.
mightyfelix
LibriVox Admin Team
Posts: 11099
Joined: August 7th, 2016, 6:39 pm

Post by mightyfelix »

Personally, I never truncate silences. There are times when I feel that a longer pause may be required. While I might trim pauses manually, I don't trust an automated process to do it for me.

To try to answer your question, I always use noise reduction first. I don't think it matters much in the long run, and some people save it for last. But I just prefer not to listen to white noise in the background while I'm making my edits.

I usually use Punch and Roll these days, as Tricia mentioned above. But not always. If I decide, for whatever reason, not to use Punch and Roll for a specific section, then when I make a mistake while recording, I stop, snap my fingers into the mic to create a loud spike that shows up well (especially when View Clipping is enabled, as Tricia also explained), and then repeat the whole phrase or sentence. It works best if I back up to the last place I took a breath or pause. Then when I edit (after noise reduction), I click forward until I find the loud spikes and cut out the bad takes.

Of course, sometimes this method means I might miss mistakes that I didn't catch at the time. One time, I said "chocolate covered Pomeranian" instead of "chocolate colored Pomeranian" and didn't notice it. I almost missed it in editing, too. It was the merest chance that I happened to notice it! So if you're worried about inadvertent mistakes, you may prefer to use Tricia's method of listening to the whole thing start to finish, correcting each mistake as you come across it.
KinoOnik
Posts: 34
Joined: July 5th, 2019, 9:53 am

Post by KinoOnik »

Hello again,

Tricia (above) said, "When I have a list of corrections to make after submission and PL..."

I refer to, "list of corrections to make after submission and PL."

Are the following steps correct/accurate as they pertain to readers?

Select a reading and get approved,
Record the selection,
Do some necessary editing (remove long pauses, clean flubs, etc.).
Noise reduction

THEN,

Submit recording for the first time, whereupon a patient PL ("Proof Listener") proofs the recording.
If I understood Tricia's comment correctly, the reader will then be informed by the PL of the additional corrections to be made marked with time stamp.
The reader then repairs the items submitted by the PL.

The Reader would then re-submit the recording for what he or she hope is accepted as the final edited version.

Do I have this straight?

Your consideration and patience is greatly appreciated. Thank you.
schrm
Posts: 4210
Joined: February 10th, 2018, 11:02 am
Location: Austria

Post by schrm »

KinoOnik wrote: March 31st, 2021, 10:42 pm Are the following steps correct/accurate as they pertain to readers?

1. Select a reading and get approved, = claim
2. Record the selection,
3. Do some necessary editing (remove long pauses, clean flubs, etc.).
4. Noise reduction and amplifying

THEN,

5. Submit recording for the first time, = uploading the recording and posting the link in the book-thread (with the duration, please :-))
6. whereupon a patient PL ("Proof Listener") proofs the recording. (most times a dpl=dedicated proof listener, one certain pl for the whole project)
7. the reader will then be informed by the PL of the additional corrections to be made marked with time stamp. = pl notes OR it can be a pl ok, also!
8. The reader then repairs the items submitted by the PL.


9. The Reader would then re-submit the recording (pls again with mentioning the duration and posting the upload link in the projects thread)
10. for what he or she hope is accepted as the final edited version. = status ready for spot pl, because the dpl will only check corrections made according to the reader. there are cases, where a dpl makes a pl note with a comment like: it is up to you, you decide. in case you feel like this is rather strict... it should be more in the direction of: working together, helping each other to get this right.

12. in case the recording is fine, it gets set to the status pl ok. (pls see point 7th, because this is the earliest point, where this status is possible).
you can see the status of your recording in four ways, by the way:
a) in the text of the forum in the pl notes.
b) in the magic window (MW), in the first posting of the book-project thread
c) on your reader page
d) in case you are a pl or bc yourself, there is a dashboard overview, which shows your readerpage-table, also


Do I have this straight? i added some numbering, comments and the abbreviations or words, which i have read most often. but: you got this perfectly right!
cheers
wolfi
reader/12275
TriciaG
LibriVox Admin Team
Posts: 60576
Joined: June 15th, 2008, 10:30 pm
Location: Toronto, ON (but Minnesotan to age 32)

Post by TriciaG »

Yes, you're right. :)

A reader should try to submit a file that they believe is ready for cataloging, so any flubs should be edited out and any adjustments made before uploading the file. The proof-listener is a safety net (to catch something you missed) rather than a crutch (to lean on for things you don't feel like looking for and fixing before submitting). ;)

Regarding schrm's post, item #4: some readers don't need to amplify. If your input volume is set correctly in the first place, you don't need to change the volume of the recording afterwards. I haven't heard your recordings to know if they're too quiet, but thought I'd mention this.
Serial novel: The Wandering Jew
Medieval England meets Civil War Americans: Centuries Apart
Humor: My Lady Nicotine
schrm
Posts: 4210
Joined: February 10th, 2018, 11:02 am
Location: Austria

Post by schrm »

absolutely, neither have i looked up your recordings!
thx tricia for the addition!
all my comments are somewhat just "the average and normal additions or used descriptions"
cheers
wolfi
reader/12275
lurcherlover
Posts: 1206
Joined: November 10th, 2016, 3:54 am
Location: LONDON UK

Post by lurcherlover »

I always proof listen to each chapter, short story or whatever, following the script to make sure the reading is 99% OK before I submit for the final PL which as has been said is a safety net just in case you have missed something. I also do not need noise cleaning and any outside noises or me tapping the mic stand with my foot are eliminated during editing. When recording I always repeat a known mistake immediately or if I have said something the wrong way or badly, and cut out the mistakes in the editing process. Each time I make a file (or a "take") I listen to it on playback, even if it's only half a page, so I can do immediate edits by making repeats of the bits I've messed up, which happens a lot. This is why an hours recorded material takes about 3-4 hours to edit.
Post Reply