This will be a detailed, inside tour of SoundStreak, as told by its CEO, Dan Caligor…and he admits, “…it takes people a while to Grok this.”
Most of you know of my interest in technological advances prompting changes in our VO business paradigms. I’ve blogged about “7 Disruptive Technologies That Have Shaped the Business of VoiceOvers“, and other product and device reviews.
Yet, one bastion of seemingly unchanging technology is the ISDN session. Source-Connect has tried, but (in my estimation) fallen short in mounting a serious challenge to ISDN…largely by failing to make the case with influential studios (see also my recent blog on ConnectionOpen).
ADVANCING THE MODEL
Now, SoundStreak is about to break on the scene with an approach that may indeed replace ISDN for ease, quality, and reliability of use. They’ll likely do it by sidestepping the entire infrastructure behind the TelCo-based (and old) technology of ISDN…and it’s much more, actually than ISDN ever offered. It’s a system…a collaboration. As Dan told me: “It’s a service, not a product”.
At it’s heart, SoundStreak makes use of the internet — but not at all in the way you might expect — to achieve it’s high-quality and reliability sessions.
First of all…let me tell you, I spent more than an hour with CEO Dan Caligor on a Skype call two days ago. No subject was off-the-table. He led me on a sample session, and openly answered all my questions. Aside from being a graduate of NYU’s film school, Dan morphed into an advisor to early-stage companies. He began to work with a man who had developed the idea for SoundStreak as a sort of academic exercise, and eventually became a partner in the start-up company that resulted. The initial vision for SoundStreak was for implementation in on-air network studios, with the implicit promise of simplicity and ease-of-use.
That original play for SoundStreak ran out of money, but now is back, stronger, broader, deeper, and more ready for prime-time than ever.
Dan himself has grown from a self-described start-up business advisor to an engineer-level technology geek fully fluent in all the idiosyncrasies of our VO corner of the Universe. He clearly has done his homework on the marketplace, the challenges of our business, and need for a convenient long-distance recording system fair to both parties.
WHAT SOUNDSTREAK IS NOT
- A costly, complicated hardware system
- Proprietary software
- Hard to install
- Confusing to operate
- A casting system (yet)
- A barrier between client and talent
SOUNDSTREAK, THE PROGRAM
Caligor stated to me: “…we are a capture and collaboration tool offered as a service, rather than as a product…” (remember: you must Grok)
Right now SoundStreak is in “private Beta”, which means you have to request an invitation or be invited by a current Beta member (there’s a few hundred right now). Currently, it’s available as a free download and will install on Mac OSX 10.6, a.k.a Snow Leopard or later. Caligor says they’re getting as many requests for older Apple OS support as they are for MS Windows support. He also says they honor anyone who requests an invite, it’s just a matter of working through the backlog.
The software must be installed on both the talent and the production (the person hiring talent) side. It’s the same application. Once installed and signed on, both parties can see each other. You are not directly connected to the other party…EVERYTHING runs through the SoundStreak servers (more on why this is important, later).
THE SESSION STEP BY STEP
- The producer begins the session by choosing a name for the session, and sends you what Caligor calls the “backing assets”. That could be a video you need to match in your narration…the script… or any other materials you need to perform your voice work for the client.
- The producer can update those materials…change, add, edit, or otherwise augment them during the session, and the changes immediately take place on both ends. SoundStreak is designed so that the talent and the production side always see and hear the same thing.
- It’s important to note that those assets are not “streaming”. After sending the materials, they reside on your computer till the session ends…so there’s never any latency when matching, say, voice to video. For security reasons, those “assets” get deleted from the production and talent computers at the end of every session…but the recording you’re about to do does not get deleted from your home computer.
- You can open multiple scripts and videos. The system keeps track of which asset you use, so when you play it back, it plays in sync with the backing asset on both ends of the session.
- Recording is done in .wav, aiff, and broadcast .wav, so you can pass-through time-coding. You can pick audio resolution all the way up to 96k, 24-bit, but the system resolves to 41.1k, 16-bit.
- The producer, then, after loading the backing asset, presses “record”. The system counts you in: 3-2-1, and you’ll read to the copy. You’ll hear it in your headphones as you read, and when you’re through, the producers presses “stop”. During the actual take, the producer listens to you over VOIP. There is a little bit of latency, according to the broadband pipe.
- Also, as soon as as soon as the in-session light goes on, a voice patch is activated, so that the production person and the talent person are talking through the system. It’s a built-in voice-patch..and it’s a smart phone patch, muting in the right order,with a talk-back button. Voice patch works between takes and from the first moment of the session to the last. When you’re not recording, it’s unmuted, but can manually over-ridden.
- At the bottom of the screen, there’s a take sheet list. There, the producer controls playback of different takes. As you do takes, each one is appended to this list as a file.
- Caligor explains: “…the whole principal of SoundStreak is that all the assets are always playing locally — both the takes, and the backing assets, and the playback is actually being synchronized through very low-bandwidth signals so it’s always optimized to work with the smallest possible amount of bandwidth…” Caligor says most of the available bandwidth is used to facilitate the voice-patch.
- There is never any compression of any soundfiles at any time during any session.
- After recording is stopped a review copy of the of the take — down-rezzed so it will transfer quickly — is sent to the Production machine within seconds. Hopefully at some point during the session (arbitrarily set by SoundStreak at 90-mins), the producer hears what he/she likes and chooses a “buy take”. Selecting that, automatically downloads the hi-res recording of that take from the talent computer — through the SoundStreak servers — to the producer’s computer. The “buy take” is uncompressed. It might take 3 to 30 seconds to download, depending on how big file is, the resolution of the file, and the fatness of the pipe connecting talent to the server to the producer.
- All during this time, all the talent has had to do is set the mic level, accept the invitation, and perform. This could seriously change the way Voice Actors record on-the-road!
- Both sides of the exchange have local copies of the takes. The talent has a hi-res and lo-res of every take (local on their machine), but not the backing assets. The producer only gets the buy-takes…and remember the backing assets are deleted at the end of the session.
- As soon as the session is over, both parties immediately receive an automatic email reporting most all aspects of the session: who’s the talent, who’s the producer, the file format, # of takes you did, start times, stop times…which take was the buy-take and more. The Session Summary email includes direct links to the archived version of each buy take on the server, so anyone you send the email to can download it directly to their machine.
SoundStreak is free right now, and probably will be for some time. Caligor says: “…the reason we’re going to keep it free for a while, is that we understand this is people’s livelihood, and as important as we think our tool is, it’s really just a screwdriver in their box, and it’s not fair to ask people to bet their livelihood on something, let alone charge them to bet their livelihood on something until they’re really comfortable with it, and therefore, we want to make that comfort level really possible easy to achieve.”
Eventually Caligor envisions a token-system of accounting. Right now, everyone gets 10 tokens, and when it gets down to one or two, you get ten more free tokens. Their plan is that most sessions, though, would be paid by the producer, in keeping with a similar protocol usually enjoyed by talent during ISDN sessions, where the producer pays for the long-distance call. One token — from either party — pays for one session, again, set as 90-minutes or less.
That every session is commissioned through SoundStreak’s servers is a plus. Caligor says: “… it allows a lot of archiving and safety and control, and for enterprise customers, it allows people to traverse corporate firewalls. That’s a huge issue at big operations. The guys in IT love us. Instead of opening up new connections in foreign places, which makes IT alarms go off…this is a single IP address…so you’re always talking to the server, not each other…”
Caligor says the entire system from top to bottom, including the server integrity is 99% bullet-proof. They’re about to end the “private Beta”, and go into a “free trial version”. They’ve never been able to crash the server, even during multiple sessions. Caligor says they run across a rare instance or two where the talent’s audio equipment presents a software driver they haven’t seen yet, but nothing that’s stopped the session. He stressed a number of times during our interview how much they’ve spent in time and engineering to make the product robust.
TESTING AND SUPPORT
The operation is lean…about 6 staffers. They subcontract out some of the development, and most everyone is taking calls and answering emails. With wider acceptance and use, Caligor anticipates more staff to handle customer support.
They want lots and lots of users to test the system. Right now, the majority of users are talent. Caligor encourages talent to tell their favorite studios about the product, and engage in tests with THEM. This is key, and may be SoundStreak’s biggest challenge: getting talent, but especially Studios and producers to figure out how to make it work for their business, and get comfortable that it’s not some kind of threat to them.
SoundStreak does require a bit of a mind-shift to understand how it differs from what’s now comfortable. We’re all busy, and we may not have the time to consider changing to, or buying into an entire new paradigm. A chicken-or-the-egg syndrome sets in. Talent won’t work with it until the studios require it, and studios won’t require it until they know talent are comfortable with it.
Caligor is confident SoundStreak is a solid product that will win over converts. “I would rather have lots of users soon, than a little or a lot of revenue soon. I’m confident that once people try it out, they’ll see its potential.”