MUSA (MUltichannel Speaking Automaton)

MUSA (MUltichannel Speaking Automaton) was an early prototype of Speech Synthesis machine started in 1975.

Audio disk containing "Fra Martino campanaro" sung by MUSA in 1978

Description

It consisted of a stand-alone computer hardware and a specialized software that implemented a diphone-synthesis technology. It was one of the first real-time TTS systems. It was able to read Italian in intellegibile robotic voice and also to sing managing up to 8 synthesis channels in parallel thanks to Linear predictive coding technology.[1] In 1978 it was released, after the building of a working prototype,[2] an 45" rpm audio disk containing some trial content of such synthesis, including the song "Fra Martino Campanaro" in "a cappella" (multiple voices) style, attached to some commercial reviews.[3] The experiment was conducted by CSELT, Turin, Italy and was led by Giulio Modena.

Notes
  1. Luciano Nebbia, Paolo Lucchini, Eight-channel digital speech synthesizer based on LPC techniques. ICASSP'79. IEEE International Conference on Acoustics, Speech, and Signal Processing. Vol. 4. IEEE, 1979..
  2. Roberto Billi, Franco Canavesio, Alberto Ciaramella, & Luciano Nebbia (1995). Interactive voice technology at work: The CSELT experience. Speech communication, 17(3-4), 263-271.
  3. (it) Virginio Cantoni, Gabriele Falciasecca, Giuseppe Pelosi, eds., Storia delle telecomunicazioni, Vol. 1. Firenze university press, 2011.

References
  • Roberto Pieraccini, The voice and the Machine, MIT press, 2015, ISBN 0262533294.
This article is issued from Wikipedia. The text is licensed under Creative Commons - Attribution - Sharealike. Additional terms may apply for the media files.