Convert Unicode file in Linux

Scenario:
          I got a file, when we open it using VI editor. i found <96> & <93> characters instead of - & "". none of the utils helped me open the file as it is.

Solution:

• Get revision / open the file in any editor,
• Copy the contents and paste it in WORD (Best decryptor of any format of content)
• Save the file as .doc,
• Open the .doc and copy the contents,
• Open the VI editor in the linux box and paste the contents and save the file
• We got the file as the orignal notations.

Issue:

      Developers used to convert the pdf to txt using some utility, due to this we can't open the file with the expected CHARSET. i've tried dos2unix, unix2dos, iconv nothing helped. :(