Louis and Bebe Barron in their Greenwich Village NYC recording studio, 1956 (photo: Walter Daran)
How to record long-distance audio interviews in a time of social distance and studio shutdowns.
Prepared for the podcasters at Teaching Tolerance (Southern Poverty Law Center)
In times like these, entering a professional recording studio isn’t smart or, often, possible. Rather than stop our recording schedules, let’s adapt. Let’s find options for recording guests remotely – to get us through current and future emergencies.
The solution must be easy to use and hard to mess up, for hosts and for guests. Audio won’t be studio quality but it must be good. (Each part of the production process — edit, EQ, mix, encode — loses a little more audio fidelity. So the better the quality of the original recording the better the audio left in the end product.)
My best cost/sound recommendations:
For host tracks, set up a simple home studio for each host.
Guests: It’s prohibitively complicated and error-prone to walk guests through all the software-install, record, and upload procedures. But it is feasible to mail guests a USB mic (or headset), with a prepaid return label. [Update coming: this is impractical in viral times. (Thx, Shea)] (If no external mic, ask if they have the wired Apple headphones: Its lavalier mic is better than using a computer’s, mainly due to mouth proximity.)
Video Conferencing Services
No: Except when unavoidable.
We moderns are all video-chat savvy. And video-conferencing services record (instructions for: Skype, Google Hangouts, GoToMeeting, Zoom). But their audio recordings are all low fidelity.
Zoom seems best for sound:
Records at 32kHz sample rate (AAC, mono, ~52 kbps bitrate). The others have a much lower 16KHz sample rate (AAC, mono, 64 kbps), so less frequency range and more chance of distortion and audio artifacts (weird noises).
Can make separate audiofiles for each user. The others mix all the voices into one file.
Gives you an audio and video file. Most others are video-only, so you need to strip the audio out.
While down this rabbit hole I discovered:
150 milliseconds [is] the maximum latency before conversations feel unnatural. Zoom works really hard to stay under 150 milliseconds, Chief Product Officer Oded Gal said. Or maybe this: Rather than optimize the connection for all devices — which means optimizing for the worst, slowest one — Zoom tends to each individually.
—Protocol
And I know people have, use, and like Skype, but: Skype requires extensive credits during the podcast (beginning, end, and every 15 minutes). While there are ways to improve the fidelity of Skype’s audio recordings, like Call Recorder (Mac) and Evaer (Win), those third-party apps still don’t match the ease, features, and quality of…
Remote Interview Audio Recorders
Yes: Whenever possible.
A recent crop of made-for-podcasters, remote-audio services record hosts and guests in different places. Two of the best are Zencastr and SquadCast (which I’m leaning toward). Both have inexpensive monthly subscriptions. Both work like this:
Starting the recording:
The host logs into the recording service and starts the session.
The guest goes to to a custom URL (sent by the host).
The interview and recording take place via the browser (i.e., no software installs needed).
During the recording:
Local recording: Full-fidelity uncompressed audio records onto host and guest computers (WAV, 44.1kHz, 16 bit).
Progressive upload: Hi-fi MP3s uploads to the cloud (so nothing lost if connection breaks).
Auto-backup: As the host/guest cloud MP3s upload, those files are simultaneously copied.
Auto-upload (local file): As soon as the host stops the recording, the WAVs (recorded locally onto host/guest computers) upload to the cloud.
Just like video-conferencing, the conversation happens over the internet (VoIP), which is bandwidth limited so loses audio fidelity. The recording, however, doesn’t use internet audio: It captures each participant’s computer input (the mic plugged- or built-in) then writes the recording directly onto their harddrives — no audio lost.
We’ll be testing the following two remote-interview recording services (and updating this post) soon. Both are well-liked by reviewers and work similarly, with a few differences:
SquadCast
SquadCast‘s advantage is their option to open a video screen (not recorded). Seeing each other speak improves the rhythm of a conversation.
They have a nonprofit discount, a free trial, but no free plan for subscriptions. Recordings upload to their cloud service. (They don’t specify where: Amazon, Azure, their mom’s basement?)
Their support docs have plenty of videos but lack screenshots: a faster way to convey computer info. But their FAQ, How to Be a Guest page, and how-to intro video are good.
Zencastr
Zencastr deserves cred for their COVID response, removing recording-hour limits on their free Hobbyist plan. Recordings upload to the user’s own Dropbox account.
They have no video screen option. (I’m a big fan of f2f interviewee interaction.) Of course, neither do tape syncs or two-way studio interviews.
Their support site is screenshot-less but has a slew of how-to videos and explainers.
Smartphone Syncs
No: Except when unavoidable.
Finally, lots of folk have posted lots of ways to record iPhone interviews (like Transom, StoryCorps, and Hearing Voices / Marketplace). But remote recording means transforming all our guests into capable recordists: color me skeptical.
Top image: Louis and Bebe Barron, electronic music pioneers, inventors, and composers of the Forbidden Planet film soundtrack (photo: Walter Daran).
The House Gates Built is not normally known as a fountain of innovation. But MS’sLiveLabs has really delivered the future in the form of Pivot. It’a web application — “don’t call it a browser”: getting lasix
“Gary Flake: is Pivot a turning point for web exploration?” ativan online
We’re navigating the web for the first time as if it’s actually a web, not page to page, but at a higher level of abstraction… So right now, in this world, we think about data as being this curse. We talk about the curse of information overload. We talk about drowning in data. What if we can actually turn that upside down and turn the web upside down, so that instead of one thing to the next, we get used to the habit of being able to go from many things to many things, and then being able to see the patterns that were otherwise hidden? If we can do that, then, instead of being trapped in data, we might actually extract information. And, instead of dealing just with information, we can tease out knowledge. And if we get the knowledge, then maybe even there’s wisdom to be found.
—Gary Flake, Technical Fellow, Microsoft; founder/director, Live Labs
Also check the TED talk for LiveLabsPhotosynth — another Seadragon derivative.
I read these things so you don’t have to: a 30-page essay, “IP Radio – A vision of radio in the Internet age” (Jan09 pdf), by Pierre Bellanger, CEO of Skyrock, the French radio & popular social net. Within are history and prognostications on all things digi-radio-future, but few new perspectives (“Radio, like the written press, the music industry and television, is changing”, etc.).
Did find a few entertaining morsels, like this (pg11) on the singularity and un-CG-ability of the human voice:
Just as a penguin can recognise another among thousands, since time immemorial we have been able to discern a wealth of incredible nuances and emotions in the human voice. We owe our survival to our brain’s ability to decipher the details of the voice, further heightening the effect of visual absence.
Pixar’s digital masterpieces such as Toy Story or Ratatouille reproduce the most complex visual experiences like wet fur or the shine of bodywork with a computer, while the characters express their emotions as well as human actors. Despite these wonders, for the voices the studio uses actors, such as Tom Hanks or Paul Newman. A voice is more complex than an image.
And this historical radio data (pg19) — unsourced, so can’t vouch for validity:
Radio experienced an auspicious period in the United States between 1980 and 2000. In 1995 radio represented a little over 10% of media advertising investments, or $12 billion. Prosperous radio stations generated results equivalent to 30% of their turnover. In 1995 regulations on ownership of several radio stations in the same market were relaxed. This 20 allowed the Clear Channel group to carry out a number of acquisitions; today, it owns about 900 stations with combined revenue of $3.5 billion in 2005.
The Clear Channel policy targeted profitability by standardising and homogenising programmes. Audiences considered risky and insolvent, such as adolescents, were abandoned. Further, morning shows, already attacked by influential puritanical groups and repeated fines from the authorities, were sanitised.
As a result, radio stations were reduced to simply playing lists of tried and tested hits aimed at an audience aged from 25 to 49 and hosted by DJ-robots. These flows were burdened with a maximum of advertising slots. Like a plane whose engine is shut off to save fuel and keeps flying for a few moments, the system seemed to work.
Then it began to show the first signs of weakness at the very moment when the younger generation was seizing the musical offer exploding on the Internet: the number of 18 to 24 year olds listening to the radio has dropped by 20% over the last ten years and 85% of adolescents now find their new music on the Internet.
In 2007 radio’s turnover was $20 billion and still represented about 10% of the media advertising market but it is a decreasing trend. For now, the years of generous cash flow are over.
However, despite everything, according to the Institut Arbitron radio audiences continue to grow (93% of the population listens to 18.5 hours a week on average!) but the length of listening time is dropping.
Finally, this possibility for targeted per-user radio ads based on IP address and other online info (pg24):
The half a million listeners who listen instantly to Difool’s morning show on Skyrock correspond to various characteristics. The same station can attract very different people which is how a large station becomes successful. The addressing process of IP radio means that people listening to the same programme can hear different adverts which correspond to their needs. Such multiple, simultaneous targeting is good news for advertisers which can concentrate their investments on suitable targets, for radio stations which can better serve their clients and are therefore more attractive, and for listeners who will hear adverts which are more relevant to what they are interested in.
IP radio combines radio’s power as a mass media with the advertising precision of the Internet.
How can an IP address be qualified, as it is occasionally random on some machines? Identification of the machine can be reinforced via a small file left on the hard disk (cookies) or through a code entered by the user at the start of the session (login) which identifies the individual. Moreover, the generalised spread of mobile terminals and their use for telecommunications will increase relevance between user and machine. This is currently the case with mobile telephones which are even more individual than PCs.
Even creepier targeted-ads could combine radio w/ the user’s web access data (pg26): “Someone listening to the radio just after looking at an automobile website could hear an ad
hoc advert via the radio.”
#pubmedia is the hashtag for Twits to spew 140char-acterizations of any public medium; so, OCPD sufferers (obsessive-compulsive posting disorder), opine away.
Matt Dunne, Google’s head of community affairs, names numbers:
Each day, Google processes 2 billion searches, sends two e-mails per second through its Gmail account system, and uploads 22 hours of video onto You Tube, per minute. It would take a lifetime to watch the video posted on the site every three months.
All these services hint at the revolutionary potential of the new computing grid and the information utilities that run on it. In the years ahead, more and more of the information-processing tasks that we rely on, at home and at work, will be handled by big data centers located out on the Internet. The nature and economics of computing will change as dramatically as the nature and economics of mechanical power changed with the rise of electric utilities in the early years of the last century. The consequences for society – for the way we live, work, learn, communicate, entertain ourselves, and even think – promise to be equally profound. If the electric dynamo was the machine that fashioned twentieth century society – that made us who we are – the information dynamo is the machine that will fashion the new society of the twenty-first century.
At work and at home, people found they could use the Web to once again bypass established centers of control, whether corporate bureaucracies, government agencies, retailing empires, or media conglomerates. Seemingly uncontrolled and uncontrollable, the Web was routinely portrayed as a new frontier, a Rousseauian wilderness in which we, as autonomous agents, were free to redefine society on our own terms. “Governments of the Industrial World,†proclaimed John Perry Barlow in his 1996 manifesto “A Declaration of the Independence of Cyberspace,” “you are not welcome among us. You have no sovereignty where we gather.†But, as with the arrival of the PC, it didn’t take long for governments and corporations to begin reasserting and even extending their dominion.
The error that Barlow and many others have made is to assume that the Net’s decentralized structure is necessarily resistant to social and political control. They’ve turned a technical characteristic into a metaphor for personal freedom. But, as Galloway explains, the connection of previously untethered computers into a network governed by strict protocols has actually created “a new apparatus of control.†Indeed, he writes, “the founding principle of the Net is control, not freedom – control has existed from the beginning.†As the disparate pages of the World Wide Web turn into the unified and programmable database of the World Wide Computer, moreover, a powerful new kind of control becomes possible. Programming, after all, is nothing if not a method of control. Even though the Internet still has no center, technically speaking, control can now be wielded, through software code, from anywhere. What’s different, in comparison to the physical world, is that acts of control become harder to detect and those wielding control more difficult to discern.
—Nicholas Carr, The Big Switch
On July 20, 1969, I mourned the last virgin moon, before man landed on it, with some hippies on a beach in Southern California. On Sept. 2, 1969, two computers exchanged meaningless data in the first test of Arpanet, an experimental military network. [transcript]
We want to thank the following companies which provide us with excellent in-kind services thru their discount and donation programs for non-profit organizations. So thanks to:
Looks like my hometown is finally letting loose it’s requirement that government job applicant’s turn over all their FaceSpaceTwit passwords, buddy lists, and secret Santa names (“Commission eliminates Facebook policy“). However, city fathers still hold onto their claim in an older policy which reserves them the right to “deflower” the first-born of any municipal employee.
For those who track copyright law, fair use, and the evolution of rights re: appropriated-cultcha and re-creation, check this TED-lecture from Larry Lessig:
No expert has brought as much fresh thinking to the field of contemporary copyright law as has Lawrence Lessig. A Stanford professor and founder of the school’s Center for Internet and Society, he chairs Creative Commons, a nuanced, free licensing scheme for individual creators.
“It’s time to party like it’s 1234567890 – ’cause it is! On this Friday, Feb 13 at exactly 3:31:30 PM (PST), Unix time will equal ‘1234567890’.”
—http://www.1234567890day.com/
Tomorrow at 6:31pm EST unix timestamp will reach 1234567890 and the nerdy will giggle with glee.
I read these things so you don’t have to: a 30-page essay, “
The 
Google Grants
VerticalResponse
Dreamhost
TechSoup
Looks like my hometown is finally letting loose it’s requirement that government job applicant’s turn over all their FaceSpaceTwit passwords, buddy lists, and secret Santa names (“