minimal usage instruction

[rom1504]

Install with pip install ClipClap

use with

import clipclap

model = clipclap.load_pretrained()
text = clipclap.generate(PIL.open("https://some/img"))

text = clipclap.generate(my_clip_embedding)

(this is an example, please tune the API as you like)

[rom1504]

https://github.com/UKPLab/EasyNMT#usage good example

[uu95]

Hello, @rom1504 do you have a pretrained model which I can use to perform audio captioning using this example?