Interface Recommendations

[Sean McCoy]

I recently set up a home broadcast studio for a client, but he's doing a lot of pre-recorded shows with call-in interviews and the inherent distortion on his voice coming back from the Telos phone hybrid is a major problem. So I've convinced him that he needs to upgrade from Sony Audio Studio to SAW Studio Basic and isolate the signals onto two tracks for post cleanup.

I need to also upgrade his audio interface from the mono-only M-Audio Fasttrack, which seems unable to record at a hot enough level, with some other stereo USB interface. He needs something that is reliable, plays nice with SAW and with a true +4 interface. Any recommendations in the $200-$300 range?

BTW, I'm also interested in how any of you broadcast-types deal with the phone interface issue. I use a simple Symetrix unit and split the signals, but it seems like there should be an easier way.

[Bob L]

The M-Audio Firewire 410 is a decent unit for about $200. The drivers seem fine and the ASIO driver can acheive low latency with SAWStudio... but the built-in headphone hardware monitoring makes it fine to use without the need for low latency.

It has 2 phantom powered mic ins and 2 separate line ins (either/or) and 8 outs.

Bob L

[Sean McCoy]

Thanks, Bob. Latency isn't an issue for him because he monitors off of a Mackie Onyx mixer pre-computer. His system doesn't have a FireWire card so I'm leaning toward USB. But if the FireWire 410 seems good to you, maybe their FastTrack Pro, which has a similar feature set, will be comparable.

[Ian Alexander]

Sean,

You didn't say which model of Telos, so you may already be a step or two beyond me here. I use a JK Audio Broadcast Host digital hybrid for phone patch VO sessions. It cost about 400.00 a couple years ago and works way better than the Gnetner analog hybrid I had before it. The JK Audio claims a null of 50 dB, while the old Gentner was about 10 dB. If I were trying the JK on air, I'd send to the hybrid before any mic compression is added. You'll probably still want to use Bob's Remove Silence on the caller in post.

Haven't been on the air in years, but I know this is still a problem for lots of stations. Way too often, you hear callers asking, "Can you hear me? I can barely hear you." I assume this is because the station has found that if they send a good level to the phone hybrid, they get too much back and it colors the sound.

HTH.

[studio-c]

A decent hybrid should be cancelling the local voice out of the caller signal. I use a JK Audio Innkeeper Digital Hybrid. I also have an old Symetrix, and a JK Audio "That2" box for a hundred bucks or so.

But here's the moneymaker secret!!!

1- Record to two separate tracks
1b- or record in stereo, host left, caller right, then duplicate the call on the multitrack.
2-Put the host signal on track 1 of the multitrack, choose mono/left only,
3- Put the caller (on multitrack track2) choose mono/right only.
4- Use the host signal to KEY the compressor on the caller signal. So whenever the host is talking, it's slamming the hell out of the caller signal, and pushing it way down.

Then you won't hear the distorted host signal thru the caller channel.

Hope that helps. I've been doing no-budget phone call programs for years. And this is cheaper than the $900 hybrid unit. Although I'm really loving that unit now. But there's always a way to get good results with little money. Somebody here can always give you a trick. And the flexibility of SS like any great console and patchbay can put any signal anywhere you want it.

Good luck!

Scott

[Sean McCoy]

We've been using a Telos One, a $600 unit and one of the current industry standards, with a trans-hybrid loss of "greater than 40 dB." For this guy, at least, I don't think anything less than 60 dB would suffice.

The splitting of the host and caller onto discrete tracks is exactly what I do and what we'll do for him once he has SSB. He is acutely aware of anything that messes with his resonant voice, so it's likely he'll prefer to manually clean the tracks. I will experiment with the keyed compressor idea, however, which could save him a lot of editing time.

[studio-c]

If the program is big chunks of the people taking turns, it's really eazy to create a view with the two MT tracks taking the full screen, and boosting the waveform so you can see who's talking at a glance. Then cutting and deleting the non talker regions. If it's a lot of back n forth, it might be a bit dicier. Still I can usually eyeball and clean up a one hour piece pretty quickly, using the k and r commands, then dragging an end if i don't like it.

[Dave Labrecque]

Sean,

I use an analog phone hybrid, which doesn't do a great job with the nulling of the not-wanted signal. The more-expensive digital hybrids are much better. When I did a radio show out of my studio for a few weeks, the folks in San Francisco who were screening the calls and putting them on the air for us were using one of the nice Telos digital hybrids. I could instantly recognize the caller sound that you hear on most syndicated radio talk shows.

If he can afford it, I'd recommend this route. It'll pay off in spades -- without all that clean-up in post being needed.

[Dave Labrecque]

A decent hybrid should be cancelling the local voice out of the caller signal. I use a JK Audio Innkeeper Digital Hybrid. I also have an old Symetrix, and a JK Audio "That2" box for a hundred bucks or so.

But here's the moneymaker secret!!!

1- Record to two separate tracks
1b- or record in stereo, host left, caller right, then duplicate the call on the multitrack.
2-Put the host signal on track 1 of the multitrack, choose mono/left only,
3- Put the caller (on multitrack track2) choose mono/right only.
4- Use the host signal to KEY the compressor on the caller signal. So whenever the host is talking, it's slamming the hell out of the caller signal, and pushing it way down.

Then you won't hear the distorted host signal thru the caller channel.

Hope that helps. I've been doing no-budget phone call programs for years. And this is cheaper than the $900 hybrid unit. Although I'm really loving that unit now. But there's always a way to get good results with little money. Somebody here can always give you a trick. And the flexibility of SS like any great console and patchbay can put any signal anywhere you want it.

Good luck!

Scott
I used this method during our radio show to eliminate a slap-back we were getting of our host (ISDN delay stuff). It worked great! (And it was all done in Live mode in SAW.)

[Sean McCoy]

He is using the Telos One digital hybrid, but it only has 40dB of separation and doesn't work much better than my 12 year old Symetrix analog unit. From the research I've done and the long conversation I had with Ted at Telos, it seems the only way to get that "pristine" talk show sound is to use ISDN for everything. Apparently they can tap the phone lines before they are converted to analog and keep the signals completely isolated that way. This client has a Musicam Roadrunner for ISDN hookups, but it, like most standard codecs, can't accept an analog phone line (POTS).

On the other hand, this guy is a radio veteran and has rightly pointed out that radio stations were able to do clean host-caller audio long before digital was around. So I'm still not clear on how all this works, but I think the perfect method would involve more expense than he's willing to absorb.

[Bill Park]

.. I use a simple Symetrix unit and split the signals, but it seems like there should be an easier way.

It is called a 'telephone hybrid'. There sed to be only a couple of makers, and they were expensive. But now, everybody makes one, there are a whole pile of them with various built in features, and they are not expensive. (But you might just get what you pay for.) A simple Mackie 1202 would be more than enough of a front end, but some of these things have that built in, too.

Bill

[Dave Labrecque]

He is using the Telos One digital hybrid, but it only has 40dB of separation and doesn't work much better than my 12 year old Symetrix analog unit. From the research I've done and the long conversation I had with Ted at Telos, it seems the only way to get that "pristine" talk show sound is to use ISDN for everything. Apparently they can tap the phone lines before they are converted to analog and keep the signals completely isolated that way. This client has a Musicam Roadrunner for ISDN hookups, but it, like most standard codecs, can't accept an analog phone line (POTS).

On the other hand, this guy is a radio veteran and has rightly pointed out that radio stations were able to do clean host-caller audio long before digital was around. So I'm still not clear on how all this works, but I think the perfect method would involve more expense than he's willing to absorb.

Sean,

I think he can get great results by simply keying a reverse gate on the caller from the host. This is better than simply keying a compressor, I think. SAW makes it all quite simple. It completely eliminates the null audio issue (and attendant distortion) by lowering the caller phone line audio (and embedded host bleed) whenever the host is talking. You hear this kind of "ducking" all the time on radio talk shows. Ever wonder why Dr. Laura never has anyone talking over her? Yeah, cuz she's a b*tch, but also because of the severe ducking that they have set up for her.

You can even adjust the floor level of the ducking if you don't like the on/off nature of full-tilt reverse gating. This is actually what I did (not compression, but reverse gating) to eliminate a bad (ISDN delay) slap of our host coming back at us on the caller line from San Francisco's hybrid. It worked like a dream.

Interesting that your guy's using the same hybrid as they were, so far as I can tell. I'd been scratching my head over why any audio from us was coming back from their "high end" hybrid (they were no help on the matter); maybe it just doesn't get any better than what we had. Absolutely not acceptable to have that echo in the host's ear. I could see getting by without the delay issue, I suppose.

Anyway, I really think this can fix your problem without big expense and without lots of manual editing after the fact. You can set it up live or set it up in post. It's up to you.

[bcorkery]

Sean,

Mix minus is the way to go. Split the host signal and record only the out bound signal. Your Hibrid should have a jack in the back that will feed out only the caller, don't send the mix to the board. I think that's where the echo is generated. The caller only needs to hear the host and the host only needs to hear the caller.

I'm using an old Digital Gentner DH-20 and it works great! With the analog Gentner, I was constantly fighting with the null and riding the gain trying to guess when the caller was going to jump in. I have no problems with the DH-20 on normal conversational levels. There's always line noise but that's minimal and there's no echo at all.

Gentner doesn't make 'em any more but the JK Audio Innkeeper box is the same thing and, as Scott said, is a steep $900. You might be able to find the DH-20 and I think they made the DH-22 also. But your Telso should be able do do the same thing. :confused:

I hope this isn't all so obvious.

[studio-c]

You have ISDN also, don't you Bill?

And yeah, I forgot to mention. I send the local feed into the hybrid thru an aux send. I also have my mic in the control room go thru that send also, so I can give instructions to the caller before we start an interview. On some podcast shows we do, we have bumper music that we roll in and do a 'live' show, so that can go out thru an aux to the caller while we record it in to SS. Then they get the feel of the whole show. But you definitely DON'T want to loop the caller signal back to the caller or you'll be fighting feedback like crazy.

By the way, for those of you working with corporate clients and/or in spoken word, a good telephone interview setup will quickly pay for itself. Especially if you get the podcast thing down. With clients tightening their belts, this is an inexpensive alternative to training videos. And it's been a great money maker for us, for not a huge cost.

When it's done, you can convert to mp3 and ftp the file to them, or host it on your own server. That's another recurring revenue stream.

Scott

[bcorkery]

Yea, I have ISDN too. It's fun when we're doing and ISDN session with the director on a phone patch. AUX definately AUX.