Welcome to the 🐸TTS project! We are excited to see your interest, and appreciate your

Thanks, <a class="user-mention notranslate" data-hovercard-type="user" data-hovercard-

Thanks for your call <a class="user-mention notranslate" data-hovercard-type="user" da

[Bug] AttributeError: 'AttrDict' object has no attribute 'generator_model' about tts HOT 2 CLOSED

coqui-ai commented on May 22, 2024 1

[Bug] AttributeError: 'AttrDict' object has no attribute 'generator_model'

from tts.

loureirorg commented on May 22, 2024 3

Thanks, @erogol. That works!

Maybe the instructions on README should be updated to avoid this kind of confusion, which I also ran into.

On the Run a tts and a vocoder model from the released model list. it instructs to use both --model_name and --vocoder_name at the same time.

from tts.

erogol commented on May 22, 2024 1

Thanks for your call @dims12 .

The problem is that you call a tts_model in place of a vocoder_model. That is --vocoder_name should take one of the vocoder we have.

Easy fix would be calling it as

$ tts --text "Это голос дикой планеты" --model_name "tts_models/ru/ruslan/tacotron2-DDC"  --out_path example.wav

then it'd use the default vocoder model set for the given tts model.

from tts.