Open tralala87 opened 1 year ago
this is because voicefixer model is really tortoisely slow to download, if you can, uncheck use voicefixer.
unfortunately it still doesn't work.. maybe I am doing the notebook wrongly.. would it be possible to make a colab for it?
Something is messed up with the argument parsing, just pass the text to stdin.
echo "hello" | ./scripts/tortoise_tts.py --voice emma --seed 42
Can't get to run the script, google colab stuck at the following step:
Downloading the main structure of voicefixer Weights downloaded in: /root/.cache/voicefixer/analysis_module/checkpoints/vf.ckpt Size: 489307071 reading text from stdin!