search

erew123 / alltalk_tts

AllTalk is based on the Coqui TTS engine, similar to the Coqui_tts extension for Text generation webUI, however supports a variety of advanced features, such as a settings page, low VRAM support, DeepSpeed, narrator, model finetuning, custom models, wav file maintenance. It can also be used with 3rd Party software via JSON calls.
GNU Affero General Public License v3.0
816 stars 91 forks source link

talk conten "" ** #184

Closed kalle07 closed 4 months ago

kalle07 commented 4 months ago

(extension in text-web-ui) is it possible to avoid talking content in "..." and in ... sometims is ok, but mostly cost time and its better i read it (that are most thoughts of the character) and better i only have the voice thats speaks to me :)

erew123 commented 4 months ago

Hi @kalle07

The only delimiter I know of to split the narrated vs spoken is asterisk *. Please see here for an explanation https://github.com/erew123/alltalk_tts?tab=readme-ov-file#-a-note-on-character-cards--greeting-messages

I can probably set content between asterisks to be dropped. Would suffice? Ive not seen anything that is starting/ending with ... before.

kalle07 commented 4 months ago

oh i meant

yes that would be fine

erew123 commented 4 months ago

Hi @kalle07

No probs. I've checked and this is already in the features request list as Silent narrator option (don't speak anything detected as narrator)

Ill be looking at it for the next larger release soon.

Thanks