Finally, introducing the Old Chinese Text to Speech Tool that I built. Ever since I took a course in classical Chinese back in 2014, I’ve felt that this has been a missing piece of the puzzle for readers of the classics.
The goal of the project is to produce an enthusiastic approximation of the language that Confucius spoke. The data set consists of 4'033 characters from the Baxter–Sagart system (William Baxter & Laurent Sagart, Old Chinese: A New Reconstruction, Oxford University Press 2014), which were available from Wikimedia.
The audio is pre-synthesized using a general Mandarin voice from Azure, which introduces problems of its own. In a later version I think I will experiment with a combination of a Cantonese and an Arabic voice to see if that can capture some phonemes better.
Obvously, this is not an accurate representation of what people talked like 2'500 years ago. But hey, there are people that believe the ancients spoke Mandarin, so I had to do something.
Send this page to you local sinologist, I would love to hear their thoughts!