Acoustic sound sources
Early on in this project I had no clear idea of which sounds or instruments would suit my conception of a speech derived music. During software development I only used very simple sonification like pure sine and noise synthesis, and later adding various spectral treatment of speech sounds as well as MIDI output for using any synthesizer be it hardware or software. What I realized though, was that ordinary synths sound like, well, synths. These soundscapes appeared to me so full of connotations to genres and musical associations that I felt I had to try to define a soundscape of my own specifically for this project. I never intended to work with the sound surface of speech or work extensively on new synthesis techniques. Computer music also sounds like, well, computer music. So in order to define a specific soundscape and also to introduce a layer of musical abstraction between the speech sounds and resulting music, I wanted to somehow bring the sound back into the acoustic world of traditional musical instruments. I first thought of robot instruments, and luckily we have a MIDI-equiped Yamaha Disklavier in our department’s studio, but building other robot instruments is still too much work and not my main focus. I then had the Idea of blending the electronic and acoustic sound realms by using contact speakers (exciters/transducers) to play back the synthesized and transformed sounds through acoustic objects like for instance musical instruments. These small devices are loudspeakers without the usual paper cone membrane, instead with the coil fixed directly onto another surface which then acts as a loudspeaker membrane or resonator. The result was a distinctively acoustical sound thus giving associations to instruments and music but still flexible enough to reveal the speech source if needed. Early testing includes mounting transducers on the sound board of a grand piano, and on different drums and cymbals to form an “acoustic drum machine” of sorts.