So, I'm going through my filing cabinet and clearing out old records. I'm following the advice of keeping only the last seven years worth of records, to avoid being buried under a mountain of old invoices and such, but mostly just to avoid having to upgrade to a larger filing cabinet or simply start boxing the older cellulose. In doing this purge, I come across the records of a really old software engineering contract.
Almost nine years ago now, in the month of September, after successfully interviewing with a certain Matthew Fleischman, I signed a contract with the games software company Mechadeus to do a brief stint of contracting. I earned about $1300 in total, at a billing rate of $25/hour. From the company's perspective, the contract was somewhat of a failure (perhaps even a complete failure - I can't really speak on their behalf). Of the time that I spent working on-site, the only product was a pile of high level, fluffed up marketing and product design ideas - with no useful software code to speak of.
Granted, stage 1 of the project - according to the contract - was to perform 40 hours of research in order to determine the technologies to be used and to scope the actual product prototype development. But stage 2 covered the development of the prototype, and this is where I flaked. In retrospect, I was definitely not up to that task. But two factors contributed to my taking the contract: 1) I did actually believe at the time that I could do what they wanted of me, and 2) I desperately needed the money. I'll leave it up to you to determine which of those factors was the overriding one.
The project was codenamed "Doppelganger". The broad concept was to enable people to essentially send talking heads to their friends in e-mail messages - most likely as MIME media attachments, but there was also discussion of tighter integration with e-mail clients. A software application would allow users to select and manipulate or design fun characters to be animated by their own voice recordings. Of course, if successfully developed, there would be a myriad other applications for the technology, with a good number of them falling inside the realm of game and multimedia production - Mechadeus' main business.
My task in this was to research voice recognition technologies that could be adapted to perform phoneme recognition from audio sample data in order to synchronize lip movement and the animation of other facial features of these "doppelganger" characters. The idea was also thrown around of offering the user the ability to mutate their voice recordings - as an example, to change the gender of their voice. TTS (text-to-speech) technology was also considered as a way to automatically translate e-mail text to audio so that no audio recording would be necessary and so that the e-mail attachments could be viable with low bandwidth Internet connections (still quite pervasive at that time). During this brief contracting stint, I was conjuring up technical terms like "Character Animation Profile", "Character Animation Script", "Voice Altering Profile", STAR (Speech-to-animation Renderer), and marketing / branding terms like "ani-mail", "see-mail", etc.
There was talk of syndicating the faces of noteworthy celebrities, such as Bill Clinton, David Letterman, Dr. Ruth, etc. There was also talk of syndicating the voices of celebrities for use in a text-to-speech engine. This is a particularly interesting idea, because at the time of this writing I am not aware of any TTS technology vendor that has marketed well known voices in the form of TTS voice profiles. The idea being that the user could get their celebrity of choice to say the things of their choosing. A "Create-a-critter" application was also envisioned, which would enable users to forge a head, so to speak. Interface elements would let users choose from galleries of facial features (noses, ear sets, eye sets, etc.) and mix and match them to create weird and whacky fun characters.
The most ironic thing about this whole episode, and the main reason I even write this account before committing all records of the event to my hungry shredder, is the striking similarity to the technology of a company that I would be working at nearly seven years later. At the time of starting employment at Sonicopia, I had no idea that this small company had recently been acquired by Pulse Entertainment, and I further had no knowledge of Pulse's business or technologies. Nor did my new employer have any idea of my involvement in this Doppelganger project nearly seven years earlier. I was hired, after all, to do DHTML/JavaScript programming for an online Web site sonification tool / service called "Sonifier".
Now, I presume that the codename "Doppelganger" derived from the idea that users might ultimately be able to create character models of themselves, to then be animated by audio recordings or TTS without the need for subsequent video recordings. This is where the similarity to Pulse's ultimate business focus is somewhat uncanny. Probably the biggest value proposition of Pulse's Veepers technology (brand name derived from v-Personality, or Virtual Personality) - and what really set it apart from its competitors in the virtual character space - was the ability to create a lifelike re-animation of a person's face from a single photograph which could then be "driven" by subsequent audio recordings or generated TTS audio. You could literally - in only a few simple steps - create your own realistic doppelganger.
But, my stint at Mechadeus was all going on way before the birth of Pulse's Veepers technology - long before Pulse started thinking of applying their vast 3D content authoring and playback framework in such a narrow and focused manner. At this stage in Pulse's history, the company was still engaged in the - at the time - very fashionable Internet content platform play, intent on creating an expansive, 3D, rich media content platform for the Internet, and competing with the grandiose ambitions of other companies such as Macromedia (of note, members of Pulse's core braintrust were themselves former "Macromedians").
It is complete coincidence that I should end up working at Pulse roughly seven years after this little blip in my career roadmap. But, it highlights the power of the Ironic Force - if you will - that is at work in our crazy universe. It's things like this, though, that also make me wonder if this life isn't just perhaps a narrative being spun by us around murky and incomplete data piped in from a greater and more complete reality, in the same way that our dreams are narratives built around the constrained data input found in that climate of sensory deprivation that is sleep - an effort to confer meaning and symbolism to sometimes spastic and chaotic renegade sensory input. OK, perhaps I'm getting too deep now. Or am I? Dreams within dreams within dreams within...