I would instead suggest using the following variation, for readability reasons mostly:
sed 's/"//g' infile
sed 's/["]//g' infile #regex variation
Single quotes escape double quotes, and vice versa. You can also enclose the character(s) to be deleted in [..]
regex character class brackets, allowing you to delete multiple characters at once.
Or consider using tr
tr -d '"' <infile >outfile
is more efficient at doing simple character substitutions/deletions than sed
, but it only works through stdin, so you have to send the output to a new file.