Using a PTH voice model

[OwOchle]

Hi, i'm trying to use a pth voice model but i get this error when trying to generate speech with it.

Traceback (most recent call last):
  File "/disks/hdd3/ai/tortoise-tts/tortoise/do_tts.py", line 39, in <module>
    voice_samples, conditioning_latents = load_voices(voice_sel)
                                          ^^^^^^^^^^^^^^^^^^^^^^
  File "/disks/hdd3/ai/tortoise-tts/tortoise/utils/audio.py", line 135, in load_voices
    latents_0 = torch.stack([l[0] for l in latents], dim=0).mean(dim=0)
                            ^^^^^^^^^^^^^^^^^^^^^^^
  File "/disks/hdd3/ai/tortoise-tts/tortoise/utils/audio.py", line 135, in <listcomp>
    latents_0 = torch.stack([l[0] for l in latents], dim=0).mean(dim=0)
                             ~^^^
KeyError: 0

Also, it's a french model, is there a way to change language in the cli or do i have to create my own generation script ?