video translator

[wangpei7001]

Thanks for the great project, I got the following error when running the video translator locally:

`

tts_models/multilingual/multi-dataset/xtts_v2 is already downloaded.
Using model: xtts
Text splitted to sentences.
['Eine meiner größten Schwächen ist, um Hilfe zu bitten, wenn ich sie brauche.', 'Ich würde das gerne besser machen.']
Processing time: 16.305320024490356
Real-time factor: 1.2662441766454846
11:41:08.189145 ERROR base_task.py:730 - Failed to convert inputs of task 'video_translation.lip_sync':
USER:AssertionError: error=Cannot convert from Flyte Serialized object (Literal):
scalar:
blob:
metadata:
uri: [...] to <class 'flytekit.types.file.file.FlyteFile'>. Expected a file, but
/var/folders/hw/b9wt5xr90gnd39nt99ckwg0c0000gn/T/flyte9qril_vy/user_space3t42q_2q/__cp is not a file. `

[kumare3]

cc @samhita-alla for the issue

[samhita-alla]

@wangpei7001 there are four files expected by the lip sync task:

unionai-examples/tutorials/video_translation/video_translation.py

Lines 252 to 255 in a873fc2

	audio_path: FlyteFile,
	pic_path: FlyteFile,
	ref_pose: FlyteFile,
	ref_eyeblink: FlyteFile,

which one of these is causing the error?

[wangpei7001]

not sure which file, I am checking the code, here is the traceback of error:

[samhita-alla]

could you share your workflow here?