As a complete newbie - re this product and synthesis in general, I have installed Synthesizer V and would like to know if there is any possibility that other languages can be used, besides the ones given.
I am interested in Hebrew, where there is predominantly one sound that does not exist in English. ie. the letter “chet” where the ch is pronounced as a guttural “G” (like when you clear your throat) - also like German ch (as in Bach).
Each voice database product has a native language (English, Japanese, or Mandarin Chinese). With the Pro edition, AI voices using cross-lingual synthesis have access to all three of these supported languages (and soon Cantonese Chinese).
Some users may use a large number of manual phoneme changes (usually implemented with a dictionary) to make a voice database sing in a language it normally cannot sing in, however this is done by using the existing phoneme list to create an approximation of a different language. Certain pronunciations will be impossible when doing this simply because the voice database cannot produce the necessary sounds for a language it does not support.
Put simply, the sounds a voice database can produce are limited by the phoneme lists it has access to.
As for the specific phonemes you mention, they aren’t included in any of the existing phoneme lists. It would probably take a lot of work to find something that sounds similar, and even then the results may only be a rough approximation.
You can find more information about phoneme entry here:
Thanks for your quick response. So, practically speaking, does this mean that I will need to wait for a voice database producer to include a language that includes this sound?
Alternatively, perhaps I could use “H” instead. This would be a weak approximation, but usable. As it happens, Americans generally cannot pronounce the guttural G - they tend to use H when speaking Hebrew.
For proper language support, yes, there would need to be a company able to partner with Dreamtonics to develop the language and voice databases for it.
The biggest obstacle here is that there’s no precident for singing synthesizers in most languages, so it’s a risky investment for a company to be the first to enter the market. Additionally, Dreamtonics is a small company, so even if they have companies willing to do partnerships to expand to other languages, they might not have enough staff to dedicate to that development effort.
But for the meantime, yes, if an accent or somewhat incorrect articulation is acceptable to you, that approach could work. You might also want to experiment with other sounds in combination with a reduced Voicing parameter.