Synthetic Voices

Modern day biological engineering and artificial life research focuses on the microscopic, the molecular, the informational, the stuff of the scientific revolutions of the past one hundred years. Our current synthetic biologies aim to turn the living into the designed, the wet into the computational, the complex into the understandable. In the 1700's, the interplay between the living and the mechanical was reversed; engineers were trying to make machines look and feel more like living things--soft, flexible, moist. Studying these historical artificial life technologies provide a valuable perspective on current synthetic biology research. Work like Jessica Riskin's "Eighteenth-Century Wetware" excellently point out the similarities and differences between the "biological" engineering of the two eras, and what this can tell us about the different philosophies that existed:

The first designers of artificial life intended their projects to resemble natural life in texture and substance, sometimes even making use of biological components. The resulting simulations, like present-day "wetware," made manifest both their makers' assumptions about the differences between animals and machines, and their impulse to undermine these differences.

Today, we struggle with finding the minimal genetic requirements for life, debating whether it will be possible to recreate all the molecular details of cells in the lab or whether it's just too complicated. The focus is molecular and informational; what is the genomic sequence that can give rise to a self-replicating cell? Conversely, how can we model molecular processes in the computer? Is it possible to make models that are sufficiently detailed enough that we can "experiment" on living systems in silico?

In the 18th century there was no understanding of life at a molecular level and engineers thinking about living processes focused at the level of human physiology. They saw bodily functions such as circulation, digestion, and reproduction, as well as human activities such as writing, playing music, and speaking as the fundamentally organic processes that should be imitated by machines. Speaking, in particular, was thought to "epitomize the organic", and for many years thought to be out of reach for mechanical reproduction. According to the Secretary of the Paris Academy of Sciences in the early 1700's (quoted in Riskin's article):

Nature had the design of placing [the instruments of the voice] altogether outside the realm of imitation. . . . Nature can use materials that are not at our disposal, and she knows how to use them in ways that we are not at all permit- ted to know.

By the end of the century, however, this impossible task was accomplished, first by Erasmus Darwin who made a wooden mouth covered with soft leather, with a larynx made of "a silk ribbon . . . stretched between two bits of smooth wood a little hollowed" able to make a few simple syllables. Over the next few decades, speech reproduction technology dramatically improved, with crazy machines made of disembodied heads able to say a few generic phrases. By the mid 1800's, Joseph Faber had designed a terrifying (at least to me) speaking machine consisting of a face that could say all vowel and consonants connected to a keyboard that could trigger each sound, allowing an operator to play a "human" voice like a piano.

At that point, however, no one cared about talking machines anymore and the whole practice of mechanical biology fell out of favor; the way that people thought about machines and life evolved. By the 20th century, the whole concept of life-like automata to reproduce biological activities was dead, replaced by the computer simulation. Now, of course, synthetic voices don't come from wood and skin and valves, but from computers.

(via rhizome)

Computer voices are becoming more and more "lifelike" as text-to-speech technology improves, with our computers and GPS thingies talking to us in increasingly "natural" voices. At the same time, our voices and our music are becoming more "unnatural" and robot-like with autotune. The complex history of the interplay between biology and engineering, living cells and machines, synthetic and natural, designed and evolved, and culture and science is fascinating and critical to how we think about new biological technologies.

(this post is inspired by and heavily indebted to Nick and his wonderful blog about sound, Noise for Airports.)