Text-to-Audio / Make-An-Audio

PyTorch Implementation of Make-An-Audio (ICML'23) with a Text-to-Audio Generative Model
MIT License
750 stars 110 forks source link

how can i use Chinese prompt? #10

Open blackwingf opened 5 months ago

blackwingf commented 5 months ago

python gen_wav.py --prompt "a bird chirps" --ddim_steps 100 --duration 10 --scale 3 --n_samples 1 --save_name "results"

when I change the prompt to Chinese, I found that it did not work

Darius-H commented 5 months ago

The dataset used for training and the pre-trained text encoder are trained on the English corpus and do not support inputs in other languages. The simple way to do this is to translate other language inputs into English.

fangg2024 commented 4 months ago

用的是英文训练的模型,我在想,是不是可以参考TTS一样,也可以用拼音注解来训练呢?不知道和底模有没有关系。