Enhancing Narration with Sound Forge 7.0

[Tristan Howard]

Hi everyone,

I'm currently working with a bunch of narration I recorded on an electro-voice 635 N/D-B dynamic omnidirectional microphone with a Sony MZ-NH1 hi-minidisc digital recorder in PCM linear mode. Anyway, so far I've only normalized my narration in Sound Forge 7.0 to -10 dB, which apparently is appropriate for speech. That helped unify the sound but my narration still sounds somewhat cold and tinny. I'm looking to warm it up and make it a bit softer by using Sound Forge but I'm not very experienced with the program. So far, it looks like I can kind of do this by going to "effects" and altering the amplitude modulation by increasing the "dry out" to about -5.6 dB. Anyway, I was wondering if anybody had any advice on how to enhance narration with Sound Forge. I'm basically looking for the sound of a laid-back guy narrating a nature documentary, sort of like Marty Stouffer used to do on "Wild America." If anybody has any input, I'd appreciate it.

Thanks,
Tristan

[Gints Klimanis]

Try some of the level compression plugins as well as bumping up some of the lower frequencies with equalization in the 250-1000 kHz region. It's really hard to offer specific advice without hearing your speech. If it sounds "dry", try adding some reverberation for small room ambience.

[Allan Black]

Does a 'bunch of narration' mean more than one person reading? Did you use an omni mic 'cause there was a few folk around the mic? How many, and male, female? You may have to process each separately to get your effect.

Yes hard to say without hearing it, but an omni-directional mic will have more room sound and less presence than the same recording made with a cardioid mic. which is the usual way to go. I'd guess it sounds too boomy, roll some bottom end out and try boosing the mids a tad.

If you can't get your result, maybe record it again with a good cardioid mic into the Sony. One extra benefit maybe a better read after hearing the omni version over and over, it always is.

Folk go, "Oh jeez not again", but maybe another hour or so and you've nailed it, versus what you've got now...forever.

[Tristan Howard]

Hey everyone,

The narration is only my own. I ended up drying it out to -4.9 dB and normalizing it to -10 dB. I think one of my problems was that my old burning software didn't encode audio too well. Anyway, I'm relatively satisfied with my narration now and will probably end up having to get a better microphone for recording speech in the future.

Regards,
Tristan

[Ty Ford]

Drying it out? What's that mean?

Omnis are seldom used for narration unless you're trying to match the sound done with the same omni used on camera as a handheld mic.

Narration is normally done with a cardioid patterened mic to lessen the sound of the room and to increase the low end of the voice.

Regards,

Ty Ford

[Paul Cascio]

I'm really interested in this topic also, but wasn't quite sure how to phrase my questions.

I'd like to know what drying out means too.

Also, when one "Normalizes" in Vegas for example, are all of the clips compared to one another, or is Normalizing done to a built-in standard?

I have problems with audible "inhales." Is a low pass filter what I need? Does a low pass filter reduce/eliminate sounds that are below a certain volume?

BTW, I don't want to edit breath-by-breath (sorry, but I just can't do it), so I'm looking suggestions on a global approach that will work on an entire recording. Am I asking too much?

Why and how is compression used on voiceover/narrative work? I recorded with a omni lav. The overall sound is pretty good but I want to learn the finishing touches.

Thanks. I've learned so much on this forum.

[Guy McLoughlin]

Originally Posted by Tristan Howard

but my narration still sounds somewhat cold and tinny. I'm looking to warm it up and make it a bit softer by using Sound Forge but I'm not very experienced with the program.

Hi Tristan,

I'm no SF expert but I generally have the most success modifying the tonal qualities of a recording using the SF Paragraphic EQ effect.

Start by selecting one of the EQ sliders, add a fairly significant amount of change ( say 12 dB ), with a wide octave spread, then run the frequency slider up and down it's range, while you monitor the effect this has.

When you start to hear something you like, modify the octave spread and or the amount of dB change, to determine how subtle the EQ can be and still give you the sound you are looking for.

Once you are happy with the sound from one EQ slider, you might want to repeat the process with another slider until you finally start to hear the sound you are after.

Often the final EQ Paragraphic filtration is subtle, but it will definitely help to fix some problems in the original recording.

If you are happy with the end result, you might want to post a few small before and after samples ( using www.mediafire.com ), for opinions from some of the expert ears here.


- Guy

[Seth Bloombaum]

Originally Posted by Paul Cascio

I'd like to know what drying out means too.

I'm going to guess here - "wet" usually means reverb, so, perhaps "drying out" might mean getting rid of reverb and room sound. I suppose the reason we've all never heard of it is that by and large it can't be done. If you're mixing with music to cover up the artifacts, sometimes a noise gate can help.

Originally Posted by Paul Cascio

Also, when one "Normalizes" in Vegas for example, are all of the clips compared to one another, or is Normalizing done to a built-in standard?

Normalizing is to a standard that typically you would set. Eg. I think Vegas has only a peak normalize function, you would set perhaps -3db, the effect would then find the peak of your track or clip or whatever, and adjust *overall* volume to set peak at that level.

The other normalize method is RMS, which uses an average volume level and sets that average to your selected volume. I don't think vegas has this, but sound forge does.

Originally Posted by Paul Cascio

I have problems with audible "inhales." Is a low pass filter what I need? Does a low pass filter reduce/eliminate sounds that are below a certain volume?

BTW, I don't want to edit breath-by-breath (sorry, but I just can't do it), so I'm looking suggestions on a global approach that will work on an entire recording. Am I asking too much?

Why and how is compression used on voiceover/narrative work? I recorded with a omni lav. The overall sound is pretty good but I want to learn the finishing touches.

A low pass filter passes low frequencies and cuts high frequencies. You're thinking of a noise gate. I find a gate occasionally useful for "breath-ing", but, where it is objectionable, usually adjust track volume envelope or cut by hand. Sorry.

Usually, narration and voiceover will be done in a studio, with the talent wearing headphones. Talent should be working to diminish breathing sounds in the mic.

Compression... I'm sure you could write a book on managing dynamic range in recording, mixing and mastering. Compression is used in a few different ways, depending on the recording, the intended distribution, and the style of the editor/engineer. The short story is that usually, if you want your voice tracks to stand out against, music, efx, or environmental noise in the playback environment most voices will benefit from compression of the dynamic range.

This is one of those times that you want really good monitoring.

There are no rules as to how much, usually you want the voice to still sound "natural", whatever that means. And one voice will be really different than the next. Mostly, you don't want to lose the quiet parts.

This is equally true of music, most pop music is mixed for playback in the car or other noisy environments, so they have very little dynamic range.

OTOH, if your program is headed for a quiet theater, it's good to have the whisper sound like a whisper.

That's an extremely general overview... note that compression will tend to enhance breathing sounds.

[Bruce Foreman]

My setup is really simple, basic almost. Pinnacle Studio 11.1 for editing and I had been using a very cheap LabTec mic from Wal-Mart plugged into the mic input on my sound card for voice over narration.

Main problem was hiss or "white noise" from the sound card at adequate mic input levels. Pulling the mic input gain down to just under 20% got rid of the hiss but left me millimeters from the mic to get useable voice levels.

Solution: An inexpensive microphone preamp, the model I ordered was Studio Projects VTB1 from zzsounds.com ($119.97). This unit has a solid state preamp stage and a 12AX7 tube preamp stage with a "blending" dial that lets you use either stage by itself or a variable blend of both. Blending in some "tube" will "warm up" cold or "dry" sounding voice some, I find a little goes a long way for me with a Sennheiser dynamic mic.

There is a line input on the front so you can run in already recorded voice from another source, phantom power switch, input gain, output gain, 70hz rolloff switch, and phase reversal. XLR and 1/4" inputs and outputs. The only thing it lacks is a headphone jack, but you probably want to do your monitoring at the piece of gear after the preamp anyway. I simply record a short test and listen to that on playback.

This unit solved my problem very well, and I really didn't need to buy the Sennheiser mic. One stereo dual pattern condenser mic lying around in my junk box since the 1980's and my Rode StereoVideoMic both work very well with the Studio Projects preamp.