synthetic speech: machinic sound poetry for the expressive masses!
mxHz, jelle dierickx, nicolas d'alessandro
monday october 17, 2005
Okno organizes a full-week workshop around the voice.
Artists of all trades are fascinated by the human voice and there is a whole history behind the use of the voice as a source for technological art: treated or synthesized, sampled or emulated, screamed or whispered, breathed or silent.
The workshop is split over a technical part, in which computer artists are introduced technically into the possibilities of the free speech synthesis software Mbrola (linux, osx, win and and running on more than 20 other systems!)
and a theoretical part, in which the focus is on the historical extended techniques for voices found in sound poetry.
At the end of this workshop a 'freestyle realtime machinic poetry' public performance is planned, in which the participants at will can battle.
Bring your laptop. We assume you are an experimental artist, and have a notion of working with audio or video, a scripting or programming language (pd/max/sc/etc).
Introduction to speech synthesis for artists:
Nicolas d'Alessandro is the designer of maxmbrola, the realtime implementation of the 'Freely Available Multilingual Speech Synthesizer' Mbrola (MBROLA reads 'EM BROLA'; think of "umbrella").
The Mbrola Project was started in the 1990ies and over the years expanded on all possible computer systems.
By now more than 60 different languages are available, including Arabic, Turkish, Polish, Lithuanian, Indonesian, Japanese, Hindi, Hungarian, Czech and Dutch, apart from English, French, German
A special focus will be on the maxmbrola objects, developed by Nicolas.
The MaxMBROLA Project
Speech and singing both result from the same production system: the voice organ. However, the signal processing techniques developed for their synthesis evolved quite differently. One of the main reasons for this deviation is: the aim for producing voice is different for the two cases.
The aim of speech production is to exchange messages. For singing, the main aim is to use the voice organ as a musical instrument. Therefore a singing synthesis system needs to include various tools to control (analyze/synthesize or modify) different dynamics of the acoustic sound produced: duration of the phonemes, vibrato, wide range modifications of the voice quality, the pitch and the intensity, etc. some of which are not needed in most of the speech synthesis systems.
A pragmatic reason for that separation is that singing voice synthesizers target almost exclusively musical performances. In this case, "playability" (flexibility and real-time abilities) is much more important than intelligibility and naturalness.
Our aim is to develop a flexible real-time application based on the MBROLA speech synthesizer allowing performers to produce complex and versatile singing - as well as speech - in many languages.
Thus, we start from a speech synthesizer and work on the adaptation of that system to real-time singing constraints. We use that particular approach for its high quality synthesis abilities.
Contemporary notes on the history of sound poetry:
Jelle Dierickx as a sound poet will introduce us into the history of the real and artificial voice. He will illustrate techniques and (im)possibilities by (non-)humans, related to sound poetry as a multidisciplinary and very diverse art form.
day 1a: introduction mbrola (concepts) (nicolas)
day 1b: introduction mbrola (practical) (nicolas)
day 2a: practical session mbrola different platforms and tools (nicolas)
day 2b: practical session mbrola speech synthesis and maxmbrola~ (nicolas)
day 3a: introduction history sound poetry 20th century (jelle)
day 3b: creative exercises mbrola (jelle + nicolas)
day 4a: creative exercises mbrola (jelle + nicolas)
day 4b: planning performance (concept and practical)
day 5a: preparation of the freestyle performance man-machine
day 5b: preparation of the freestyle performance man-machine
+ evening public performance...