What should be the input shape of this?

[Hassibayub]

Hey Experts,

I started working on my project of choice after a deep learning course predicting words from a sequence of images.
I collected the sequence of 15 images of (128, 128, 3) shapes that make up one of the words "Adam", "Daniel" (The best teacher🙌), and "Frank" (totally random words).

I made a pandas data frame of two columns images and target where one images cell/element contains the list of numpy array of 15 images [arrary(128,128,3), array(128,128,3)...] and the target is the word ( Adam, Daniel, Frank).
Now I want to apply an LSTM/GRU model because it has a sequence, but I am confused about what should be the input shape? length of the list is 15
should the input [ ]?