I know. I know because I’ve used many sounds to replicate foreign languages and it’s something I’m aware of.
I played the other way around, shorter u and longer U. But for this word you also have to play with the pitch curve that has to rise from a lower pitch to the correct note for the u.
What you also have to notice is that also English words do not sound equally, so why do we suddenly are expecting this from a new implemented language?
Example, the “oo’" in these two English words sound differently: book and boot. And my other examples.
I do not expect words to sound equally.
My issue is not: words I expect to sound equally sound different
it’s: words that should be phonemized with the same symbol (even if they sound different) are phonemized with different symbols.
also “book” and “boot” have different vowels. book is with uh and boot with uw.
I am not expecting all German words to sound the same. I am well aware of the nuances of German spelling:
Weg [ve:k], weg [vɛk]
Drache [ˈdʁaxə], Sprache [ˈʃpʁaːxə]
Buch [bu:x], Fluch [flu:x] are the same sound BUT Bruch [bʁʊx]
schick [ʃɪk], Sick [ziːk] (as a surname with irregular spelling)
I am not saying “all German words should sound the same”. English words don’t sound the same, neither do French words, or Italian, or Portuguese, or even Chinese. I am learning Chinese so I know that.
2 Likes
isn’t U supposed to be the short sound and u the long sound?
another bug spotted
These characters are completely silent, but they should have a sound
I also notice SynthV tries to nasalize phonemes even in languages that don’t support them, though it might not always sound correct.
sometimes get weird results like this making a sh sound but I’m surprised it doesn’t just… declare it invalid and move on
hey, this sounds nasalized… somewhat?
and this… not so much
This Spanish sound seems to make an ы or 으 sound
I’m aware these are training accidents but I’m surprised how much you can squeeze out from this
this is just a smudge sound. but then…
these long-lost phonemes work again.
Sticky jju is no more!
but then…
it turns into an English approximant r sound.
the i~ in Spanish makes a German like ü sound
also I can use this suffix in two variants, with and without the underscore, which yield very slightly different sound outputs
the one with the underscore sounds like the slender R in Irish
Yes but U has the short but open vowel sound , while u, even though it is the longer variant, sounds like a closed vowel and therefore sometimes it is better to shorten it by closing with U, without having to try for a long time with the parameters of the Mouth opening. This works well, for me at least.
YES!!! I use this parameter. I use it for different sounds. I do practically the same - if a particular vowel doesn’t work I often replace it with a different one and use mouth opening
1 Like
Here’s my Wishlist for the final release:
-
Add German, French and Portuguese to more voicebanks.
-
Realistic TTS [Text to Speech] talk mode [including the voicebanks you don’t have] with free unlimited text to speech voicebanks on the Dreamitonics website.
-
Returning mode from SynthV 1: Manual Mode
-
Country Pronunciations [Liam Exclusive]
-
European Pronunciations for Portuguese
-
Canadian Pronunciations for French
-
Create a voice [similar to Vocoflex but standalone]
-
Master vocals and music automatically with Master Ai
-
Generate an ai melody and chord from lyrics
I would add:
- Return the
|a|button from v1
Using the technical preview version with Cubase 15.0.30 causes crashes. Vocals are not generated properly. I reverted to the latest version, 2.2.1, but now the vocals cut out periodically. I’m extremely disappointed.
Please create a tool to clean up all the changes made by the technical preview version as soon as possible. As it is, I’m afraid it will break my past projects, making the DAW unusable.
I hope for a fix as soon as possible.
I can’t enter spaces without having to turn off a setting, but it resets after some time, it’s very annoying. Also, certain symbols don’t work. very annoying bug.
When will the beta release outside Dreamitonics?
About This Technical Preview
This release differs from our standard beta program. We are currently evaluating several major changes in parallel, which have not yet reached the feature maturity typically required for a beta. However, given the scope of these updates, early feedback is highly beneficial. This allows us to validate our direction and make informed decisions before committing further to specific development paths.
Why would a third party wish to invest money in changes to their voice banks if those ‘new features’ could becom unuseable if certain experimental features are not used in a possible future beta - never mind the stable release?
Would like to try the pre beta called 2.3.0.tp1. But as mentioned below this requires to replace my V2 production installation at my risk which I do not want. See e.g. Bitwig how this can be handled in having the beta versions in parallel to the productive one. Something like that could make me try the dreamtonics pre beta as well. Otherwise I will have to wait until the stable version is available.
1 Like
What are you even talking about?
And installing and using a technical preview, as with beta versions, is a very deliberate and conscious decision.
I don’t understand what do you mean.
What they mean, is that they want an separate installation : one for the stable release, one for beta/technical preview. they don’t want the beta to replace the stable release installation.
2 Likes
I found a bug:
The French language eats up any word ending in a consonant, when followed by a direct phoneme input
Hello, I just bought a windows laptop that uses ARM. With the beta download link listed here, do you need to have the purchased software to activate it? I downloaded the demo but it is not compatible. Should I just wait for the ARM support to fully release?