Babble es un sistema para crear poesía sonora automática, inspirado en el uso de vocablos para describir sonidos musicales, jugando en el espacio entre el discurso y la percepción sonora.
Sound poetry is poetry free from the usual contraints of syntax and semantics, where only the sounds of words are used to form the structure and feel of the poem. The Ursonate by Kurt Schwitters is one fine example.
Vocable words are used by musicians to describe the sounds their instruments make, while teaching, swapping tunes and during performance. In many cultures formal systems of vocable words have developed, including the bol syllables of the Indian tabla drums, and the canntaireachd of the Highland bagpipes.
Comparitative research into the perception of sound and speech has found both differences and correspondences. That there are two conflicting modes of perception is made clear by the phenomenon of sine wave speech. On the other hand scientific research such as that carried out by Aniruddh Patel shows correlations between the rhythm of music and speech in different cultures, as well as between neurological activations of the listeners of music and speech.
Babble sits on the boundary between speech and music, allowing play with the structure and form of simplified phonetics without having our perception hijacked by speech processing.
Under the hood is a physical model of a guitar string (Karplus-Strong synthesis), and a formant filter to produce vowel-like sounds. The different consonants control the excitation and noisiness of the string, and the vowels control the tuning and the formant filter. The result is something like a speech synthesiser, but with the vocal apparatus largely replaced with a single guitar string.
Although babble’s phonetic system is quite unlike that of English or any natural language, when faced with the text that generates the sound, as well as the vowel-like queues, the listener has the sensation of ‘hearing’ speech as the brain searches for meaning.
Leave a Reply