Why does NPR sound so unique?

February 19, 2007 1:33 PM

Why do people's voices on NPR sound so dramatically different from those on any other radio station?

I might be crazy, but whenever I hear someone talking on an NPR station I can instantly tell that it's an NPR program; it's just something about how the person's voice sounds so intimate and soft spoken. So assuming that this isn't just my own ears, what causes this effect? Is it post production, an NPR "manual of style" that dictates whispering into a microphone a quarter inch from your mouth, or perhaps exclusive use of some "just you and me in this quiet log cabin" brand microphones? Both real answers and "you're just crazy" replies are welcome. Thanks!

posted by kaytwo to technology (33 comments total)

No, you're not crazy. I would consider it just as I consider sportscasters (or Valleygirls or hiphoppers, etc, etc..). They listen to it, they aspire to be it, they become it.
posted by HuronBob at 1:45 PM on February 19, 2007

You are not crazy. The New Yorker once described the voices on NPR as "mentholated" which is the perfect way to describe the NPR voice.
posted by psergio at 1:48 PM on February 19, 2007

It's not just you. SNL's old sketch The Delicious Dish parodies those exact voices.
posted by muddgirl at 1:49 PM on February 19, 2007

I've heard NPR uses very expensive Neumann microphones, but I'm sure that's just a part of the NPR Sound.

One downside to The Sound: Daniel Schorr slobber. Yikes.
posted by marionnette en chaussette at 1:54 PM on February 19, 2007 [1 favorite has favorites]

Not in the USA and have never listened to NPR, but I suspect this could be similar to the Radio 4 effect here in the UK. The voices sound fuller, closer and richer than on other stations.

Do you find that you have to turn the volume up louder on NPR than on other stations? My hunch is that the effect you're hearing is probably because NPR is using less compression in their production.

Your average sports or pop music station compresses the audio signal so it sounds better on cheap radios (gross simplification). But radio stations of the NPR/Radio 4 ilk, and classical music stations, use less compression in order to allow for a fuller dynamic range, and, arguably, truer audio reproduction.
posted by chrismear at 1:57 PM on February 19, 2007 [3 favorites has favorites]

You're not crazy at all, but I think it's a much a matter of what they refuse to do as it is a distinctive style in itself.

They don't use compression effects to the same extreme as on most commercial radio, and reject that "announcer voice" style that is so common in advertisements and with commercial DJs.

Tom McCourt wrote an interesting dissertation about NPR at the university of texas called "National Public Radio and the Rationalization of the Public." In it he talks about the shift in the early 1970s from Educational Radio to Public Radio as we now know it. There was a 'folksiness' in the original mission that worked to contrast commercial radio. This can now still heard in Prairie Home Companion, and, despite all the changes in the format through the years, their relaxed vocal tone might be seen as a vestigial remnant of this founding ethos.

A senior NPR producer was quoted in McCourt as saying--near the beginning of Public Radio--that “as much as people need to hear two more reports about Eastern Europe or the third world, they also need to hear about how to pick rhubarb for a pie," referring to this combination of hard news and 'down home' cultural programming.
posted by umbú at 1:59 PM on February 19, 2007

NPR Reporter's Equipment. Microphones and recording equipment can also play a part in creating a particular sound.
posted by tumble at 2:00 PM on February 19, 2007

I think it is a combination of the technical factors above plus NPR and its on-air talent making an effort to sound sober and rational as opposed to the frothing at the mouth of Rush Limbaugh or Randi Rhodes.
posted by TedW at 2:08 PM on February 19, 2007 [2 favorites has favorites]

I suppose I'm in the minority here, but that "NPR voice" is distracting to me. I always chalked it up to the talent being mostly writers who happen to read their stuff on air rather than people who have good radio voices being picked to report news and stories.
posted by monkeymadness at 2:14 PM on February 19, 2007

Excitement sells. Commercial radio announcers like to keep the excitement level up. They talk faster and consistently louder, which, even if your volume is down, still sounds more energetic. The engineer tries to keep the output level in the upper end as much as possible. (See how much real silence is in a typical radio or TV ad. usually even what seems to be a pause has a background noise come up to maintain the volume level).

NPR has no need for that kind of false excitement. Presumably, the listeners listen for the content.
posted by ctmf at 2:25 PM on February 19, 2007

In short, it's pretty much been said - it's attitude and production value.

NPR, and the programs that originate there have something of a unified sound - They all either come from NPR headquarters in Washington DC, or at NPR West, in Culver City, CA. Studios in a building tend to be set up the same way. This is for both audio consistency and ease of maintenance.

They definitely try to avoid the "Commercial Sound," and one way they do this is by being pretty delicate with the compression. Having listened to many the NPR program straight off the satellite, and looked at more hours than I can imagine in Adobe Audition, they're just not as compressed. In fact, they're hardly compressed. That can change, however, at your local station. My station is pretty well compressed, but still not to the extent that my previous employer (an AM-Talker that syndicated ol' Rush himself) was. I mean, commercial stations just crush their signals, and that's unheard of in NPR programming.

As for NPR using Neumanns? Possibly. There's a lot of great vocal mics out there, but I haven't seen their studios. A lot of depends on context - you can't use 4 or 5 mics like that in a room full of people. It'd be chaos. So when Neil Conan has a bunch of guests on Talk of the Nation, they're probably using something else. A mic like a Neumann would be better for the newscasts, and the newsmagazine shows.

And Daniel Schorr's mouth noise? I'm actually more distracted by Carl Kasell's. Oh man...
posted by god hates math at 2:41 PM on February 19, 2007

I am a public radio broadcaster and host segments on a public radio show.

It's not that the people on public radio are "mentholated" or that they speak differently from other Americans, nor is it down to what microphones they use.

No, it would be criminal for a host to modulate their voice electronically. Imagine meeting Ira Glass in person and discovering he really sounds like Bob Edwards. I do sound like Bob Edwards, although my mother says I sound like the continuity announcer at KQED (I think that's a compliment).

In 2001, Mark Goodier of BBC Radio One moved from that station to Classic FM and newspapers noted how badly he suddenly sounded on the new station. He had mixed his own voice at Radio One but the equipment to do that wasn't with him at Classic.

The truth, and be ready for it, is that people who host NPR's flagship programs are there because listeners want to hear standard american voices on the air. If ATC suddently got a host with wide Minnesota vowels, complaints would erupt.

Although I thought my voice was great, I have had to spend about three grand on coaches to show me how to use my voice more effectively for public radio audiences even though I really just aspire to be an executive producer.

The other thing is that until not too long ago, many of the people who did public radio for a living were volunteers or badly paid - many still are at public radio stations around the country. Those with great, really good voices get a gig going very quickly and then never let it go. For most including me, it's not about the money, it's about the love for the show.
posted by parmanparman at 3:00 PM on February 19, 2007 [4 favorites has favorites]

As to Chrismear, the compression theory is bollocks. I worked at two of the BBC's national networks and they only ever sounded good when the hosts bothered to show up early enough to learn about what they were going to talk about.
posted by parmanparman at 3:04 PM on February 19, 2007

I was interviewed by remote once at an NPR station. Sitting alone in a tiny studio, headphones on, waiting for my slot, I put my page of notes down on the desk and.... swoooOOOSH! The soft glide of paper meeting laminate. Wow, did that sound nice, and loud, too. I did it again.

I've done a lot of radio, but never have I sat in front of as sensitive a mic. I didn't notice the make, though, sorry.
posted by Scram at 3:07 PM on February 19, 2007

You would probably find This article, entitled "The Darkroom Magic of NPR", to answer most of your questions.
posted by melorama at 3:16 PM on February 19, 2007 [5 favorites has favorites]

Personally, I think NPR's "media sausage" style of editing to be fucking annoying as hell. It's like the talk radio equivalent of a Kenny G or Steely Dan album: Technically perfect, but devoid of any soul or spontaneity.

And I say that as a rabid fan of many NPR shows.
posted by melorama at 3:18 PM on February 19, 2007 [2 favorites has favorites]

I was waiting for Parmanparman. I didn't think he'd be arguing with me, though. It's alright though - the next time I engineer his show, I'll make sure hears the difference...

Kidding!@
posted by god hates math at 3:22 PM on February 19, 2007

parmanparman, are you saying that what I explained about compression is simply not true, or are you just saying that it's not responsible for the effect kaytwo is talking about?
posted by chrismear at 3:30 PM on February 19, 2007

I've always noticed the same effect of which you speak. I don't know that compression is the answer, as I have little expertise in audio engineering. But, I've always attributed it to a more relaxed, less gimicky, timbre and delivery; and less processing of the signal.
posted by Netzapper at 4:00 PM on February 19, 2007

The $3000 mics do help, but only in the studio, and it's not like they're exactly uncommon elsewhere. Not compressing the hell out of a signal does significantly improve the sound (at the expense of relative, apparent volume), but as someone almost said above, that's at the mercy of local affiliates in NPR's case.

Perhaps it's just that the programming is more painstakingly made. Outside NPR, I wonder how many other sources of speech radio there are in the US which undergo any production at all?

(Stayed put then, parmanparman?)
posted by genghis at 4:15 PM on February 19, 2007

[Oh, and from a personal point of view I know how to make myself sound better. But then I come from a UK and strictly outside-the-BBC background where everyone drives their own desks.]
posted by genghis at 4:18 PM on February 19, 2007

As a former NPR host (for a local station in Iowa) I figure I should weigh in.

The NPR voice is, for the most part, a practiced cadence. It's quite different from a normal speaking voice, though I wouldn't call it artificial. Even now, I can impress friends by announcing the weather and upcoming programs on the fly.

That aside, there are a few tricks. Most every radio studio I've worked in tweaks the equalization on the microphones slightly. The bass and mid-tones are emphasized while the nasal highs are taken down a bit. Add in a pop filter and it all comes together to create a richer and slightly warmer sound.

The key is to keep it simple and in moderation - otherwise the sound is, as evidenced by the case parmanparman mentioned, artificial and strange.

And don't forget that most radio studios are padded with sound dampening foam. They're quiet places that minimize echoes. That alone does wonders for the overall sound.
posted by aladfar at 4:29 PM on February 19, 2007 [4 favorites has favorites]

Great question. I blogged about this a few months ago:

Dear NPR:
I humbly request that your on-air personalities and guests do one or all of the following:

1. Use a wind-screen on the microphone.
2. Have a glass of water nearby.
3. Speak at least five inches away from the mic.

This would apply especially to Carl Kasell who, at times, sounds as if he’s choking to death on his own saliva. Peter Overby sounds like he’s just swallowed a whole lemon sprinkled with alum. Other guests speak so close to the microphone that I can hear the inner-workings of their digestive system. This is not compelling radio. Its a gross-out contest and you are winning.

It freaks me the fuck out. Please, stop it!

I love NPR but with the treble all the way down.
posted by KevinSkomsvold at 4:39 PM on February 19, 2007 [4 favorites has favorites]

Not NPR but for what it's worth the speech on the Australian station ABC Classic FM always sounds very nice to me. There's a very rich velvety tone to it which I don't recall hearing on, for instance, BBC Radio 4. In some way that's difficult to describe it sounds as if the announcer is speaking quietly very close to you.

I've always assumed it's something to do with the way the studios are set up. I think in some contexts it might sound a bit overdone but given the job of station is to play 'classical' music it seems appropriate it's done that way.
posted by southof40 at 4:45 PM on February 19, 2007

Definitely second the "darkroom" article that melorama linked to. Be sure to listen to the audio as well (from NPR's On The Media.

They don't really talk about the voices so much as about everything else that makes NPR sound unique. On a related side note, for years I wondered who the guy was who makes all the sponsorship announcements, his name is Frank Tavares.
posted by timelord at 4:49 PM on February 19, 2007

Don't know if it is for the reasons set forth above, but I have always believed that Bob Edwards is a terrible radio voice. His stuff is often interesting, and I particularly like his radio interviews with musicians on their new releases, but his delivery is just not good for radio. He has too much of a tendency toward conversational speech patterns, and tends to fade out and then come back in again.
posted by yclipse at 5:39 PM on February 19, 2007

Oh my god Carl Kasell is awful. I can't listen to him his marble mouthed mushiness and lip smacking is so freaking distracting and disturbing that I can't even pay attention to what he's actually saying. When he comes on, I change the channel.

Proof positive that NPR people are not chosen for their radio voices.
posted by jaded at 3:48 AM on February 20, 2007

Cool question and answers. I have always wondered by NPR sounded so different.
posted by octothorpe at 3:55 AM on February 20, 2007

As to the effect the OP described, I always thought it was largely affectation/mannerism. One thing I've noted among at least some of the hosts is what sounds like a deliberate attempt to remove many of the orthodox stresses and rythms from speech -- so everything sounds vaguely interrogative, or maybe aimless.
posted by Clyde Mnestra at 7:07 AM on February 20, 2007

Wow. Ask never fails to impress the pants off of me. Thanks everyone for your amazing answers!
posted by kaytwo at 11:30 AM on February 20, 2007

I see nobody's mentioned the fact that NPR people often sound as if they're aiming their remarks at young children with comprehension deficiencies. Pauses between words and phrases, artificially heightened changes in pitch, desperately chirpy intonations in general.... To tweak the immortal words of Dorothy Parker, Tonstant Wistener fwowed up. (And yet I listen—partly because my wife loves NPR, partly because it's the most intelligent radio available in this neck of the woods.)
posted by languagehat at 11:47 AM on February 20, 2007

Confession: I registered vowell.org circa 1996 hoping it would somehow lead to Sarah contacting me.

Didn't.
posted by iconjack at 12:39 PM on March 20, 2007

I have long noticed this and have a few ideas:

I once knew a guy who for decades was the voice of Proctor and Gamble. He told me that their sound was achieved using ribbon microphones and a slight tape speed reduction on playback.

Among other factors, such as speech patterns, and a forced intimacy vocal style, the uber realistic sound may be achieved using Blumlein Pair mic setup.

I have listened for audio processing / sweetening artifacts, related to digital effects processing / compression, and I can't discern any, if there are any, its a very light touch.

There does seem to be a overall equalization profile for all of the programming content that boosts the lows and mids a touch.
posted by walkerc at 10:12 AM on August 17, 2007