Editing spoken audio
Editing spoken audio
I'm in the process of putting together an audio workshop. I'm looking for a reference (web site, book, etc.) on how to go about creating and editing spokent audio in general. I'm familiar with most of the features in Goldwave that I need, it's more the general sequence of events. For example, if I make a mistake recording and need to do a re-take, is it better to wait until I'm done to do it, or do it as soon as I notice and insert it later? What do I listen for when editing? What is the general sequence to go through in creating audio?
Any guidance would be appreciated. Thanks.
Al
Any guidance would be appreciated. Thanks.
Al
-
- Posts: 550
- Joined: Wed Dec 01, 2004 11:06 pm
- Location: Denver, Colorado
I do voiceovers, so I may be able to give some suggestions that you will find helpful.
I do not record the entire project as one long continuous track - in most cases - because some projects have been long enough to completely fill two full CDs. What I do is keep an ever lengthening "Project" file and a separate "Workparts" file. I open a new file - my "Workparts" file and record as much as five minutes on it and then copy and paste that to my "Project" file. I only do this once I have listened critically and am sure it is what I want to save in the "Project" file. I usually edit out the flaws on this relatively small file because it does not put at risk the rest of the project file in the event of some catastrophic screwup.
When recording voice only projects, you will end up with occasional little things known as "mouth music". I prefer to edit out this unwanted noise at this point - it works better for me this way.
These unwanted noises, clicks and ticks and little pops won't be there consistently because many factors influence the production of these unwanted noises. The best thing to do is avoid them to the degree possible. Drink plenty of water - dehydration is one cause. Another factor is condition of the internal mouth parts - tongue, palate, cheeks, lips etc. What you have eaten or have drunk may help or hurt you here. Apple slices are good. Soda pop is bad, for instance.
I prefer to record at a relatively low level, the average level staying at around -6 or thereabouts, which allows a little "headroom" for those instances where certain sounds may be louder than the average of the track - recording at a higher level may cause distortion in those particular transient sounds. The volume can always be raised later on, using the tools available in Goldwave.
I do not record the entire project as one long continuous track - in most cases - because some projects have been long enough to completely fill two full CDs. What I do is keep an ever lengthening "Project" file and a separate "Workparts" file. I open a new file - my "Workparts" file and record as much as five minutes on it and then copy and paste that to my "Project" file. I only do this once I have listened critically and am sure it is what I want to save in the "Project" file. I usually edit out the flaws on this relatively small file because it does not put at risk the rest of the project file in the event of some catastrophic screwup.
When recording voice only projects, you will end up with occasional little things known as "mouth music". I prefer to edit out this unwanted noise at this point - it works better for me this way.
These unwanted noises, clicks and ticks and little pops won't be there consistently because many factors influence the production of these unwanted noises. The best thing to do is avoid them to the degree possible. Drink plenty of water - dehydration is one cause. Another factor is condition of the internal mouth parts - tongue, palate, cheeks, lips etc. What you have eaten or have drunk may help or hurt you here. Apple slices are good. Soda pop is bad, for instance.
I prefer to record at a relatively low level, the average level staying at around -6 or thereabouts, which allows a little "headroom" for those instances where certain sounds may be louder than the average of the track - recording at a higher level may cause distortion in those particular transient sounds. The volume can always be raised later on, using the tools available in Goldwave.
Recording & editing practices
Thanks so much for the advice. I really appreciate the technique of the "Project" and "Workparts" file. That makes a lot of sense.
Al
Al
-
- Posts: 550
- Joined: Wed Dec 01, 2004 11:06 pm
- Location: Denver, Colorado
Here are some websites you may find useful or interesting:
http://www.vo-bb.com
http://www.commercialvoices.com/June_06.php
http://www.peterdrewvo.com/html/home_vo ... tudio.html
http://www.vo-bb.com
http://www.commercialvoices.com/June_06.php
http://www.peterdrewvo.com/html/home_vo ... tudio.html
Retakes
I find that if you make a mistake while recording, the best thing is to take a couple of seconds to compose yourself, and then record the line again. Then at edit time you can cut out the mistake.
I find it's best to do the retakes straight away (some people call these "pick ups"), because you have more chance of your voice having a consistent tone. If you do pick-ups at the end, you often find there is a subtle change in pitch or pace that makes the speech sound a bit weird when you have finished editing.
I find it's best to do the retakes straight away (some people call these "pick ups"), because you have more chance of your voice having a consistent tone. If you do pick-ups at the end, you often find there is a subtle change in pitch or pace that makes the speech sound a bit weird when you have finished editing.
-
- Posts: 98
- Joined: Mon Apr 10, 2006 1:37 pm
- Location: Jersey
-
- Posts: 550
- Joined: Wed Dec 01, 2004 11:06 pm
- Location: Denver, Colorado
When recording a "pick up" or patching a mistake it is usually more desirable to record that correction or replacement line immediately so as to avoid voice mismatches. The resonance, timbre and tonality of the voice does move around a bit and coming back to replace or repair something later will often result in having a voice timber that is sharp and resonant contrasted against a drop in that is perhaps a little softer or huskier or thinner, etc.
Another good technique is to replace an entire line as a discrete unit instead of trying to record just one or two words and expecting them to fit smoothly into the surrounding context. Sometimes that may work - but not often in the real world. It's kinda like repainting that dent in your car door once it's been fixed - the painter doesn't just paint a six inch circle. He masks off an entire area which allows the paint to appear more even and properly matched instead of leaving a big ol' bullsey on your door.
Another good technique is to replace an entire line as a discrete unit instead of trying to record just one or two words and expecting them to fit smoothly into the surrounding context. Sometimes that may work - but not often in the real world. It's kinda like repainting that dent in your car door once it's been fixed - the painter doesn't just paint a six inch circle. He masks off an entire area which allows the paint to appear more even and properly matched instead of leaving a big ol' bullsey on your door.
-
- Posts: 98
- Joined: Mon Apr 10, 2006 1:37 pm
- Location: Jersey
^^^^spoken well this is very true.....
...if you mess up, just repeat the sentence.....
Also, depending on how you record...or how LONG your recording will be.....sometimes, in a controlled env, i will make two quick taps on the mic.....perhaps with my nail, or a pencil or something. This just puts two large spikes in the visual audio, and when editing, it allows you to jump right to the mistake.
...if you mess up, just repeat the sentence.....
Also, depending on how you record...or how LONG your recording will be.....sometimes, in a controlled env, i will make two quick taps on the mic.....perhaps with my nail, or a pencil or something. This just puts two large spikes in the visual audio, and when editing, it allows you to jump right to the mistake.
-
- Posts: 550
- Joined: Wed Dec 01, 2004 11:06 pm
- Location: Denver, Colorado
Togglehead - that tapping idea is a good one. I usually do something similar. I whistle very loudly - no, that is not the visual clue that locates the edit point - what happens is that the pain in me eardrums usually makes me scream out in agony and I usually end up smacking the side of my head into the mic - of course the hard plastic of the headphones cracks and the prospect of spending another $12.95 for a set of high dollar cans gets me to cussin' real loud.....
And THAT"S the visual cue that locates my edit point.
And THAT"S the visual cue that locates my edit point.
-
- Site Admin
- Posts: 4372
- Joined: Wed Mar 10, 2004 6:43 pm
- Location: St. John's, NL
- Contact:
-
- Posts: 550
- Joined: Wed Dec 01, 2004 11:06 pm
- Location: Denver, Colorado
-
- Posts: 98
- Joined: Mon Apr 10, 2006 1:37 pm
- Location: Jersey
oooo good one don....actually i got the idea from something very similar.....i was recording a VO track once in a quiet booth and the engineer turned up the can volume and burped in to the Talk back.....scared the ever living CRAp outta me and i smacked into the mic.....BINGO....perfect edit point.
Chris are you saying do the Ctrl+Q while recording? Usually in my field of working with telephone recordings, i dont get to actually record the file.....i just get my hands on the "finished" file.....
....s'why i love the right and left click selection method so much.
Chris are you saying do the Ctrl+Q while recording? Usually in my field of working with telephone recordings, i dont get to actually record the file.....i just get my hands on the "finished" file.....
....s'why i love the right and left click selection method so much.
-
- Site Admin
- Posts: 4372
- Joined: Wed Mar 10, 2004 6:43 pm
- Location: St. John's, NL
- Contact:
-
- Posts: 98
- Joined: Mon Apr 10, 2006 1:37 pm
- Location: Jersey