Multichannel Speaking Automaton

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

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] a 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.^[4]

• Eloquens (software)

• Pieraccini, Roberto (2012). The voice in the machine : building computers that understand speech. Cambridge, Mass.: MIT Press. ISBN 978-0262533294.

• "MUSA" (in Italian). Archivio Storico Telecom Italia. Archived from the original on 2018-01-02.
• Baggia, Paolo (2006). "W3C: Speech Technologies and Standards - DIT Seminars, Povo, Trento" (PDF). W3C (in Italian). Retrieved June 22, 2022.