• Joined on 2023-03-13
nk990 commented on issue mrq/ai-voice-cloning#133 2023-03-19 14:50:33 +00:00
Mispronouncing certain letters for slavic languages

@psammites yes it's correct

So before a vowel is it labiodental? Or is it bilabial and part of a dipthong?

phonemize and the SAMPA included in SOFES (converted to IPA) don't…

nk990 commented on issue mrq/ai-voice-cloning#133 2023-03-17 15:50:29 +00:00
Mispronouncing certain letters for slavic languages

Before a vowel, the pronunciation is labiodental, [ʋ].[5] Before or after a vowel, the pronunciation is bilabial [u̯] and forms a diphthong.[5][13][14] At the beginning of a syllable, before…

nk990 commented on issue mrq/ai-voice-cloning#145 2023-03-17 10:32:54 +00:00
Utilities>Tokenizer does not work after TTS is loaded

I saw you started to train the Slovenian model, any results, could you share some examples?

nk990 commented on issue mrq/ai-voice-cloning#133 2023-03-17 00:41:09 +00:00
Mispronouncing certain letters for slavic languages

@mrq I suggest to change istead of:

phonemes = phonemizer( text, preserve_punctuation=True, strip=True ) if you could try this:

` phonemes = phonemize(text,language=lang,strip=True,…

nk990 commented on issue mrq/ai-voice-cloning#133 2023-03-17 00:27:10 +00:00
Mispronouncing certain letters for slavic languages

@nk990 Do you have an IPA-annotated Slovnenian dataset? I added the missing symbols to models/tokenizers/ipa.json but SOFES is transcribed in SAMPA and the UCLA Phonetics Lab Slovenian corpus is…

nk990 commented on issue mrq/ai-voice-cloning#133 2023-03-16 08:42:43 +00:00
Mispronouncing certain letters for slavic languages

With the default tokenizer, yea. It just strips out all the accents. That's part of the reason @nk990 is running into issues, to it there's no difference between [č] and [c] (or [ć], or [ĉ],…

nk990 commented on issue mrq/ai-voice-cloning#133 2023-03-15 15:29:08 +00:00
Mispronouncing certain letters for slavic languages

Post-nap, post-coffee edit: Replace [č] with [q], [ž] with [x], crank up the Text LR Ratio to the max and let her rip. Should work without replacing the tokenizer or diffusion model.

this…

nk990 commented on issue mrq/ai-voice-cloning#133 2023-03-15 08:41:25 +00:00
Mispronouncing certain letters for slavic languages

At last, I can train it to speak Ubykh (or at least pronounce გვფრცქვნი)!

@psammites Can you also share you experience with us? Like how many clips do you have in your…

nk990 commented on issue mrq/ai-voice-cloning#133 2023-03-14 18:58:58 +00:00
Mispronouncing certain letters for slavic languages

Nope, I used default config settings. Gonna try to increase text LR ratio, can you also share your config file that you've been using for japanese, just for comparison of the parameters.

nk990 opened issue mrq/ai-voice-cloning#133 2023-03-14 08:57:33 +00:00
Mispronouncing certain letters for Slovenian language
nk990 closed issue mrq/ai-voice-cloning#130 2023-03-14 08:28:43 +00:00
Error while runing training
nk990 commented on issue mrq/ai-voice-cloning#130 2023-03-14 08:28:16 +00:00
Error while runing training

Your copy is a couple of days out of date.

Right, after getting latest version everything works fine, thnx for your great job!

nk990 opened issue mrq/ai-voice-cloning#130 2023-03-13 15:39:51 +00:00
Error while runing training