Klatt’s history of speech synthesis: A Few Highlights

A few of the most interesting clips of synthetic speech in Dennis Klatt’s archive.

 


The VODER, the FIRST EVER electronic speech synthesis, demonstrated at the New York World’s Fair, 1939, by Bell Laboratories. (Archive #1)


The only MONOTONE SYNTHESIZER ever made! Designed to make research on speech acoustics very easy. Yet, the man-in-the-street assumes computers talk with flat intonation. The ‘ Pattern Playback’ contributed greatly to our understanding of speech acoustics. Haskins Laboratories, New York City, 1951. (Archive #2)


First song in synthetic speech, “Bicycle Built for Two” with synthetic piano. This song was reprised by the `decorticated’ Hal in `2001: A Space Odyssey‘. (This is the bonus clip, and was not part of Klatt’s collection.) Bell Laboratories, by Louis Gerstman and Max Mathews, 1961. (Not part of Klatt’s Archive)


Can you tell the REAL from the ARTIFICIAL? These short phrases were copied from natural recordings by John Holmes in 1961. It is quite hard to tell the real from the synthetic. However, weeks and weeks of labor was required to produce each phrase. (Archive #7)


SPEAK-N-SPELL toy. The first mass produced high-tech speech product. The system announces words to be spelled using a simple keyboard. Popular for 3-8 year olds in the early 1980s. (Archive #13)


MODERN SPEECH SYNTHESIS BY RULE: MITalk / DECtalk. Dennis Klatt devoted many years to development of MITalk, a research system that converted ordinary printed text into intelligible speech synthesized entirely “by rule”. (Archive #33) The commercial version of this system, DECtalk, comes standardly with the following voice options:

A. ‘Perfect Paul’ — talking at about 300 words/minute (Archive #36)

B. ‘Beautiful Betty’ (Archive #35B)

C. ‘Huge Harry’ (Archive #35C)

D. ‘Kit the Kid’ (Archive #35D)

E. ‘Whispering Wendy’ (Archive #35E)


MODERN SPEECH SYNTHESIS BY RULE: VocalTractLab. Peter Birkholz created a free, interactive multimedial software tool to demonstrate the mechanism of speech production. The source code and additional information can be found here. The source website has other examples:

Dona Nobis Pacem (audio only)