There is work to create a synthesizer for Udmurt speech to date. It was reported by the press office of the Udmurt Federal Research Center (UdmFIC) of the Ural Branch of the Russian Academy of Sciences.
Work continues on filling the corpus of the Udmurt language. Several functions are already available.
Now our main task is to fully automate the work, so that the program itself checks for errors, converts words and introduces them into the corpus when entering text
The corpus is filled with 55 thousand words with explanations and examples of usage.
When the dictionary is completely filled, it will be possible to create interactive services for translation and dubbing of texts in Udmurt.
We hope that in a few years we will have a translator with neural network technology – when the text is entered entirely, say, in Russian, and the program translates into Udmurt
Let us remind you that the courses of the Udmurt language will resume from August 3.