This post is not about the content of talk radio programs, but about the sound of talkers on talk radio, especially those not in the studio where the program originates.
This will not be an extremely complex technical article (I hope) though some technical material needs to be presented in a simplified (perhaps over-simplified) way. If readers desire a later post that goes into a lot more technical detail, email me or leave a comment in the comments section of this post.
The “Olden” Days
When I was growing up, through my high school and college years, there was no use of digital audio. There were experiments and research, but to the best of my knowledge none of these were commercially available.
If a radio station wanted to broadcast the play-by-play of the local high school basketball team’s games, or a concert given by the local symphony, they generally rented a “Class A” phone line which could carry speech and music with quality that at least sounded decent on AM radio, and some even on FM (which was just coming into play). This was simply a copper cable but one that didn’t bandwidth limit the signal and was supposed to also be “clean” of background hiss or static (though in my experience was not always that good). This was still an analog connection, one designed for reasonable quality and offered at a pretty steep price.
Home phones also used analog copper wire technology in those days (some still do today though that technology is dying out) but each line was limited to a very narrow frequency range giving the typical “telephone” voice sound. By limiting the bandwidth the phone company could use multiplexing techniques to allow more than one call to share the same physical cable. The goal of that telephone technology was to provide good intelligibility (for speech) even though the fidelity was weak. The limited bandwidth also filters out some extraneous noise.
As music programming gradually moved to FM, which provided very good fidelity, many AM stations switched to “talk radio” often involving listeners calling in to ask questions or, more commonly, give their own opinions (sometimes at length). But the voices of those callers, even though usually intelligible, were grating on the ears of some listeners such as me. If someone was calling in to ask a short question or to provide a short piece of information, it seemed OK, but some stations programmed hours of over-the-phone voices. It did make the airwaves available to the public. A good cause.
Stations that needed to send their program signal to a transmitter located a few miles away (perhaps on a mountain) could rent the pricey Class A phone line from the telephone company though some used microwave transmitters and receivers for that purpose. A good short distance solution. But none of those high quality sound technologies were really available to the average listener. A radio news reporter describing a story “on the scene” over the air sometimes also had to use a low quality telephone line (such as a pay phone) though later, as solid state technology developed, some manufacturers made portable transmitter/receiver systems that were acceptable for such jobs and always available, thouogh still not the highest voice quality, (from my limited experience with one from a manufacturer I worked for). Major networks or large stations could ultimately use satellite technology, but, as most people know from listening, they can suffer from quite a delay between when an anchor starts asking a question and when the remote reporter hears it and answers. This delay is caused by the signal having to travel each way to a distant satellite then back to earth. Also, using that technology “on the scene” required a truck with a satellite dish attached. Expensive and bulky (though still common today for television reporting where good picture quality, as well as sound, is needed).
The Big Change
Digital technology changed the way audio could be transmitted. Early on, radio stations could have a permanent news program originating in say, Philadelphia, to be received and aired on a station in, say, Albany, NY. The secret was a ISDN (Integrated Services Digital Network) line. These were expensive compared to consumer phone lines, but cheap enough that professionals could afford the system. Even a voice-over talent, who needed to voice many commercials, could own an ISDN line and feed his/her voice to a radio station without having to go to the studio. Thus a voice-over artist in San Francisco could be hired to voice a spot for a station in Detroit and the audio quality would be good (provided the voice talent had a pretty good voice studio in his/her home or personal studio).
I should explain briefly how audio gets converted to a digital bit stream. An analog-to-digital converter “samples” the audio signal every fraction of a second (there are different standards ranging usually from 32,000 samples each second to 128,000 or more). The sound level of each sample is converted into an integer and forms the basis of the audio “bit stream.” Although the full bit-stream is captured faithfully if one is making a audio CD, it usually requires too much bandwidth to be used directly over the internet. Thus comes into play the “codec” (short for “compressor-decompressor) which digitally compresses the digital bitstream (often by throwing much of it away but hopefully in a way that is not so noticeable to most people). The decompressor and digital-to-analog converter changes the compressed data stream to digital audio. For remote radio the analog-to-digital system and digital-to-analog may be part of the same “black box” as the codec.
ISDN technology used slow digital lines compared to today’s lines, but because they used a very robust (though less efficient in terms of bandwidth) circuit-switching technology rather than packet switching, that system, now 30 years old, is very stable for voice and there is very little delay when conducting a two-way interview, so the interview can sound pretty much the same as if the two participants were in the same studio. Makers of ISDN audio systems, many still in use today, didn’t always use the same codecs, so if a remote broadcaster didn’t have the same brand of transcoder as a radio station carrying his/her broadcast, sometimes the signal had to pass through another service that could change codec formats. An ISDN line cost several hundred dollars per month and the transcoder, a one-time purchase, could be in the thousands. Fine for radio professionals to use, but not accessible to the average citizen who wants to have his/her voice heard on the radio.
The bad news, for audio professionals, is that ISDN is not very profitable for the phone companies and many simply won’t install new service. They want to get rid of ISDN. And although practical for remote interviews and voice-overs, it is not a technology which could be adopted to the non-professional. But for radio, it is reliable and high quality (though not suitable for video).
Enter the New Contender
What changed the audio world fast was VOIP (Voice Over Internet Protocol). The internet is a packet-switched network — so there are issues to be discussed below — but it is available cheaply. I expect that most of the readers of this blog have used VOIP even if they didn’t realize it. When one orders a bundled service from a cable company that includes internet access, home phone, and cable TV, the home phone service is VOIP. It does not use the traditional analog copper wires from your home to the phone company. In fact the “home phone” offered by your cable company is really a joke because it costs very little to use VOIP over the internet. My Blackberry PRIV Android cell phone will switch to VOIP when I am connected to a WiFi router. Thus at home I can talk on that phone for hours without using any of my cell phone “minutes.” By charging for “home phone” service a cable company can say “Look how cheap we are, just $29.95 (or whatever) each for cable TV, internet. But notice that there is always the disclaimer “when bundled.” You cannot get just cable TV and internet for $29.95 each if you do not take the phone service “bundle.” But remember, the phone service costs the cable company almost nothing. My Blackberry works as a VOIP phone over WiFi and would continue to work that way even if I canceled my cell plan or my “home phone” service from Verizon Fios, as long as I keep the high-speed internet.
VOIP TALK RADIO
The big question is how VOIP technology can be useful for remote interviews, call-in talk radio, linking remote news radio news operations to a home station, etc.
Do VOIP calls have better sound quality than traditional analog phone lines? The answer is generally “yes,” though various factors can influence both. One limiting factor for home call-ins is that usually the caller is using a traditional phone whose microphone may not be up to higher quality standards. Also, if one is calling in from a cell phone connected to a cell network, one has to consider the quality of that network. Many of my friends have commented on how much more natural my voice sounds when I use my Blackberry over VOIP then when I call over the T-Mobile network. To make the sound quality better on VOIP I often use a fairly pricey Danish headset with microphone when using my Blackberry. If I did a radio interview using my Blackberry and VOIP via my WiFi it would not sound nearly as terrible as old fashioned analog “talk radio” or as a call-in using my cell network (as those nets usually limit voice quality for various reasons). I would not sound like I was in the studio with the program host, but my voice might not sound as grating to the listener. The advantage of most VOIP technology is the fact that it is accessible to anyone cheaply. Radio station WNYC in NYC does a 60-minute call-in segment every Friday with the mayor of NYC. He often seems to call in using some kind of VOIP technology. His voice is clear but still not a natural as if he went to the station.
Regular reports, interviews, etc. (that are not “call-in”).
Suppose WNYC wanted to do a better quality broadcast (from the sound standpoint) every Friday so it would sound like the mayor was in the WNYC studios. An ISDN transceiver could be installed in the mayor’s office and another one at the station (which I expect they actually have). But we have noted that ISDN is expensive and being phased out.
Packet switched internet audio can suffer from the following issues:
- Because packets are routed differently along the internet there can be a time delay (much like satellite) that makes an interview difficult (if done live rather than being recorded and edited before broadcast). This latency can be hard to predict.
- Internet “glitches.” VOIP is not as robust as ISDN in terms of not dropping, garbling, or delaying audio. This may not always be a problem but I have participated in internet conferences where the connection would quit, requiring reloading of the software.
- If a station wants the highest quality there are hardware and software that offers better than consumer phone quality, but often it means the same hardware and software must be used at both ends of the conversation. Not only are different codecs sometimes incompatible, some offer varying quality.
- Internet data speeds vary. Most people have fast broadband connections but before one says “I have way more capability than audio needs” one has to consider how many other users may be using that same connection at the same time.
Solutions to consider:
One popular VOIP solution that offers better quality than a regular VOIP phone is Skype. This has the advantage that anyone can easily download the software so it is easy for a subject of an interview to connect to a radio station via Skype. The BBC uses it quite extensively. A second advantage of Skype is that it can be can have bideo capability so it can be used in television interviews. But even for radio it is nice to see who you are talking to and get visual clues that may help you do a better interview.
The user has little control over the Skype audio codec (at least that I can find) so you have to take what they give you. This is usually much better than other consumer call-in solutions. Even though the sound is not usually “studio quality” many users make it worse through the following bad habits:
- They talk from a room with poor acoustics. A room without much reverberation and without hard walls is best.
- They use a low cost microphone, often the one built into their computer.
- They don’t talk close enough to the mic. If one is using the mic built into one’s laptop computer then it is hard to sit close enough to it to still see the screen. The solution here is to use a headset (as I do). But I don’t mean a $29 headset from an office store. Good voice quality requires something better. Mine cost $200.
These suggestions apply to nearly any kind of voice interview or speech of course, not just to Skype.
But there are even other problems with Skype:
- It does not allow for overlapping conversations, so if one person interrupts another it simply cuts off the person talking.
- This might change, but it is my current understanding that the Skype license agreement for radio broadcasting requires that the station state it is using Skype and must be identified as a Skype call at least every 15 minutes. [A great way for Skype to get free advertising. Imagine if this was a requirement for other broadcasting technology: “This interview is conducted using Sennheiser microphones . . .”]
Of course, Skype may well be in the process of upgrading its software and services and, as I stated earlier, the BIG advantage of Skype for many applications is the video capability. It is the “picture telephone” envisioned, but not perfected, for many decades. I use it for social calls to friends (nice to see their faces) and, when I worked at a university, we often conducted initial job interviews via Skype for candidates who did not live near our city.
Are there other solutions?
There are but these require both parties to a conversation to have the same software. Not a problem, as we mentioned, if a radio station is doing regular broadcasts from a specific remote location but not really good for “call-in” shows or very occasional interviews. A station cannot have every piece of software made in case someone calls in with an uncommon one. You can search professional broadcast station suppliers for hardware/software solutions. There is not enough space for me to cite many examples here.
Since I expect a number of readers of this blog work on low budgets but want to do real time podcasting or low power radio, I want to highlight one product that seems to offer much promise: Feenphone.
Their website is https://feenphone.com/
This is free software though they solicit donations. It is not clear how well the company is doing but with their software being free there is not much risk in trying it. It offers better quality than Skype and can allow overlapping conversations. They have several examples on their website. Their motto is “Our mission to improve independent spoken media.” It is well worth visiting their site. I have not, at this point, used their software so I am not really a “promoter” of their product. But it does interest me. The user at each end has to use a computer (laptop or desktop) running the Feenphone software. As with all broadcasting (as mentioned above) a good microphone and audio interface on each computer will produce a professional sound. The Feenphone software provides the codec.
If any reader is using any of these or other solutions to remote interviews for radio or podcasts please feel free to describe your system in the comments section of this post.