Giulio Modena
Giulio Modena is an Italian engineer who led the development of MUSA at CSELT in Turin in 1975, the first stand-alone speech synthesizer able to read text in intelligible Italian. Born in Pontelongo in 1940, he graduated from the University of Padua with a thesis on voice formant extraction and joined CSELT in 1967 to head its Electroacoustics section. A 1978 version of MUSA could sing a cappella, and a 45 rpm record released in 1978 carried its Fra Martino Campanaro. From 1979 he headed the User Terminals Division, and in 1988 he coauthored a paper on CCITT standards for digital speech processing. His group's speech work, which grew to include speech recognition, later moved to the spin-off Loquendo, which became a company in 2001.
Timeline
-
By 1967
A Padua thesis on formants
He is born in Pontelongo, Italy, in 1940. He graduates from the University of Padua with a thesis on voice formant extraction.
-
1967
Electroacoustics head at CSELT
In 1967, CSELT in Turin hires him to head its Electroacoustics section. There he researches electroacoustic measurements, electroacoustic transducers and speech signal processing.
-
1975
MUSA, a speaking automaton
In 1975, he leads the CSELT voice technology group that builds MUSA (MUltichannel Speaking Automaton), the first stand-alone speech synthesizer able to read text in intelligible Italian. The system runs diphone synthesis on its own hardware and is one of the first real-time text-to-speech systems.
-
1978
MUSA learns to sing
A second version of MUSA, released in 1978, can sing, including polyphonic a cappella pieces, and a 45 rpm record from 1978 carries its Fra Martino Campanaro. With MUSA, CSELT becomes in 1978 the only industrial group besides AT&T with speech synthesis technology of industrial significance.
-
1978–1980
An artificial speech signal patent
In 1978, he and Aldo Reolon publish "A new artificial ear for telephone use" in The Journal of the Acoustical Society of America. With Stefano Sandri and Carlo Scagliola, he invents a CSELT method for generating an artificial speech signal to test telephone transmission equipment objectively, and the US patent is published in 1980.
-
1979
Head of the user terminals division
From 1979, he heads the User Terminals Division at CSELT.
-
1986–1991
Papers on speech coding standards
In May 1986, he presents "Speech Processing: an Outlook Esprit Founded European Research" at the International Workshop on Recent Advances and Applications of Speech Recognition in Rome. With Maurizio Dècina, he publishes "CCITT standards on digital speech processing" in the IEEE Journal on Selected Areas in Communications in February 1988. In February 1991, he coauthors "Speech and image coding for digital communications" with Mario Guglielmo and Roberto Montagna.
-
1990–2001
The Loquendo spin-off
His CSELT voice technology group, which also works on speech recognition, publishes Advanced Algorithms and Architectures for Speech Understanding, edited by Giancarlo Pirani, with Springer in 1990 together with the consortium of an ESPRIT project. The group's work later moves to the spin-off Loquendo, which becomes a company in its own right in 2001.
Sources and credits
Photo: We have no published photo of Giulio Modena yet. Send us a free-licensed photo.