Making a speech synthesizer
Hello.
I would like to talk about a very relevant topic.
There are speech synthesizers. Look, how many English synthesizers there are. Lots of.
We have recently created a new Russian synthesizer for RHVoice.
Of course, it's useful and necessary, but unfortunately, not relevant.
Who will use the new Russian voice? Agree, move from one voice to another is a difficult and long process.
But the essence of my post is not a criticism of what is being done. But a request for help in creating something that doesn't exist.
Have you ever wondered that there are languages for which there are no speech synthesizers?
Or is there, but they are so poor quality that it's better not to use?
I would like to give a specific example that prompted this letter.
Uzbek language.
What is there?
Espeak with Uzbek voice. Yes, it is. But ask a native speaker of Uzbek how it reads. Answer: Bad.
What is being done?
Unknown.
My friend who is a native Uzbek speaker says he contacted the creator of RHVoice, Olga Yakovleva, about creating an Uzbek voice for RHVoice.
But alas, there has been no news for a long time.
I learned articles by programmers who had experience in developing speech synthesis. As a result, from a development point of view, the development of the voice engine is a laborious and difficult task.
Therefore, the only option is to use ready-made tools.
And here is a question for those who have learned this topic.
What can we do ourselves? Without the help of truly professional specialists? And can we at all?
So far, I see it as a single option, to modify the existing Espeak voice.
But maybe someone can suggest another solution?
And after all, Uzbek is not the only language that needs a speech synthesizer...
Thanks in advance!
-- Audiogames-reflector mailing list Audiogames-reflector@sabahattin-gucukoglu.com https://sabahattin-gucukoglu.com/cgi-bin/mailman/listinfo/audiogames-reflector
