WaveNet: text to speech using a generative deep learning model. Existing text-to-speech systems use parametric generation or a concatenative approach where tiny samples of a recorded voice are strung together to create synthesized speech. Using a deep learning technique WaveNet generates synthetic speech a single sample at a time. Especially interesting: "If we train the network without the text sequence, it still generates speech, but now it has to make up what to say. As you can hear from the samples below, this results in a kind of babbling, where real words are interspersed with made-up word-like sounds" The academic paper.
Wendy Carlos is one of the most important composers living today. While primarily connected to the fields of electronic music, sound design, and alternate tunings, her compositions transcend these genres. It is certain that her music will be included among the major milestones of 20th century music.
Vowels can be tricky to describe phonetically because they are points, or rather areas, within a continuous space... [more inside]
Much easier to put together custom DNA An interdisciplinary study led by Dr Ali Tavassoli, a Reader in chemical biology at the University of Southampton, has shown for the first time that 'click chemistry' can be used to assemble DNA that is functional in human cells, which paves the way for a purely chemical method for gene synthesis. Writing in Angewandte Chemie International Edition Dr Tavassoli's team and his collaborators, Dr Jeremy Blaydes and Professor Tom Brown, show that human cells can still read through strands of DNA correctly despite being stitched together using a linker not found in nature.
In an age where composite video devices are oft relegated to thrift stores, select videographers are embracing the format and its analog roots via video synthesis. [more inside]
Only rarely is there an opportunity to participate in a molecular 'happening'. On an open field at Stanford University in 1971, several hundred students convened to undulate and impersonate molecules undergoing protein synthesis by a ribosome. Narrated by Nobel laureate Paul Berg and performed by a cast of very groovy cats. (via)
I just finished up reading The Turk by Tom Standage (briefly mentioned in passing here) a biography of the chess-playing automaton that toured Europe and later the Americas during the pivotal transition from the 18th to the 19th century. The Automaton was invented as an exercise in national pride by Wolfgang von Kempelen, who considered it a trifle compared to his experiments with mechanical speech synthesis. As a celebrity, the automaton had historic encounters with Benjamin Franklin, Napoleon, Beethoven, Philidor and Charles Babbage, and fictional encounters with the monarchs Catherine the Great, George III and Frederick II. Standage credits it with influencing the development of the Difference Engine, the power loom, Poe's mystery stories, and Barnum's manipulation of the press. The myths surrounding have even caught James Randi, who seems to have been unaware of a colleague's reconstruction based on notes from the last owner.
Physical synthesis models vibrating structures to synthesis sounds (Tao home). SAOL lets you program audio synthesis inside an MP4 file (MP4 home).