Editing Wolfgang von Kempelen's Speaking Machine (1769) (section)

==== 1. Studies on the Reproduction and Invention Process of the Machine: ====
While Kempelen's work "Mechanism of Human Speech" appears to provide a fairly detailed description of the appearance, illustrations, and the quality of synthesized speech of his final version speaking machine, there are numerous ambiguities and inaccuracies in these descriptions, casting doubt on their ability to faithfully replicate the machine's original form.

For instance, Kempelen's use of the Latin alphabet sometimes results in unclear associations between letters and phonetic qualities. For example, the distinction between whether the letter "O" represents a semi-open or semi-closed rounded back vowel remains unclear. Additionally, the reason behind the absence of front vowels between "E" and "A" is ambiguous. Kempelen's use of "CH" for palatal, velar, and possibly uvular voiceless fricatives ("higher" and "lower Ch") further obscures the specific "CH" phonetic qualities that the speaking machine can synthesize. Moreover, Kempelen's claim that opening only one "nostril" of the machine can synthesize an /n/ phoneme, although intriguing, lacks conclusive evidence.

And for the evolution of Kempelen's invention process. The question of why Kempelen modified the machine's construction after publicly demonstrating a seemingly convincing version, and whether these modifications aimed to simplify the machine to align with his work "Mechanism of Human Speech," remains unanswered.

Besides that, Kempelen introduced a "membranous pipe model" in his work "Mechanism of Human Speech," suggesting that it could simulate human vocal cords. However, he chose a distinctly unnatural model for the machine's glottis. This decision presents a puzzling question for future scholars to delve into.<ref>Brackhane, Fabian. “DIE SPRECHMASCHINE WOLFGANG VON KEMPELENS – VON DEN ORIGINALEN BIS ZU DEN NACHBAUTEN,” n.d.</ref>