in R, table() is used to build contingency tables as specified in R doc : https://www.rdocumentation.org/packages/base/versions/3.6.2/topics/table

table uses the cross-classifying factors to build a contingency table of the counts at each combination of factor levels.

To build an actual table, or in the correct nomenclature a "dataframe" use data.frame() function Also rename the variable accordingly

count_table <- data.frame(seurat_clusters = Sample@meta.data$seurat_clusters, 
                               orig.ident = Sample@meta.data$orig.ident, 
                            singlr_labels = Sample@meta.data$singlr_labels)

Then plot

ggplot(count_table) + 
  geom_bar(aes(x=singlr_labels, fill=orig.ident), position = "dodge") + 
  facet_wrap(~seurat_clusters)

But you could directly use seurat metadata and plot it without building a temporary dataframe as :

Sample@meta.data %>%
  ggplot() +
  geom_bar(aes(x=singlr_labels, fill=orig.ident), position = "dodge") + 
  facet_wrap(~seurat_clusters)

Advice : Try to take a R basic course to familiarize with R syntax. Once that learned you will be much more confortable working using R.