diff --git a/Hs-conllu.cabal b/Hs-conllu.cabal
index 85f2c0f..12a28fd 100644
--- a/Hs-conllu.cabal
+++ b/Hs-conllu.cabal
@@ -17,6 +17,6 @@ library
   exposed-modules:     Conllu.Type, Conllu.Data.Tree, Conllu.Data.Token, Conllu.IO, Conllu.Parse, Conllu.Diff, Conllu.Utils
   --other-modules: Conllu.Utils
   -- other-extensions:
-  build-depends:       base >=4.9 && <5, containers >0.5.8 && <0.6, directory >=1.3 && <1.4, filepath >=1.4 && <1.5, parsec >=3.1 && <3.2
+  build-depends:       base >=4.9 && <5, containers >0.5.8 && <0.6, directory >=1.3 && <1.4, filepath >=1.4 && <1.5, megaparsec >=6 && <7, void <1, text >=1 && <2
   hs-source-dirs:      src
   default-language:    Haskell2010
diff --git a/src/Conllu/IO.hs b/src/Conllu/IO.hs
index a80e00d..5ff6f4b 100644
--- a/src/Conllu/IO.hs
+++ b/src/Conllu/IO.hs
@@ -10,17 +10,17 @@ import System.Directory
 import System.Environment
 import System.FilePath
 import System.IO
-
-import Text.Parsec.String
+import qualified Text.Megaparsec as M
 
 ---
 -- uses customized parser
 readConlluFileWith :: Parser [Sentence] -> FilePath -> IO Document
 readConlluFileWith p f = do
-  r <- parseFromFile p f
+  d <- readFile f
+  let r = M.parse p f d
   case r of
     Left err -> do
-      print err
+      putStr . M.parseErrorPretty $ err
       return $ Document f []
     Right ss -> return $ Document (takeFileName f) ss
 
diff --git a/src/Conllu/Parse.hs b/src/Conllu/Parse.hs
index ac9008f..ff0b273 100644
--- a/src/Conllu/Parse.hs
+++ b/src/Conllu/Parse.hs
@@ -1,3 +1,4 @@
+{-# LANGUAGE OverloadedStrings #-}
 module Conllu.Parse where
 
 {--
@@ -21,11 +22,13 @@ import Data.List
 import Data.Maybe
 import System.Environment
 import System.IO
+import Data.Void
+import qualified Data.Text as T
 
-import Text.Parsec hiding (token)
-import Text.Parsec.Combinator
-import Text.Parsec.String
-import Text.ParserCombinators.Parsec.Char
+import qualified Text.Megaparsec as M
+import Text.Megaparsec.Char
+
+type Parser = M.Parsec Void Stream
 
 ---
 -- conllu parsers
@@ -33,7 +36,7 @@ document :: Parser [Sentence]
 document = documentC sentence
 
 documentC :: Parser Sentence -> Parser [Sentence]
-documentC s = endBy1 s blankLine <* eof
+documentC s = M.endBy1 s blankLine <* M.eof
 
 blankLine :: Parser ()
 blankLine = litSpaces <* newline -- spaces shouldn't exist, but no
@@ -43,7 +46,7 @@ sentence :: Parser Sentence
 sentence = sentenceC comment token
 
 sentenceC :: Parser Comment -> Parser Token -> Parser Sentence
-sentenceC c t = liftM2 Sentence (many c) (many1 t)
+sentenceC c t = liftM2 Sentence (M.many c) (M.some t)
 
 comment :: Parser Comment
 comment = do char '#'
@@ -59,8 +62,8 @@ tokenC :: Parser Index -> Parser IxSep -> Parser Form -> Parser Lemma
   -> Parser Token
 tokenC ix is fo l up xp fe dh dr ds m =
   mkToken <$> ix
-  <*> optionMaybe is
-  <*> optionMaybe ix <* tab
+  <*> M.optional is
+  <*> M.optional ix <* tab
   <*> fo <* tab
   <*> l  <* tab
   <*> up <* tab
@@ -76,11 +79,11 @@ emptyField = do char '_'
                 return Nothing
 
 index :: Parser Index
-index = do ix <- many1 digit
+index = do ix <- M.some digitChar
            return (read ix :: Index)
 
 indexSep :: Parser IxSep
-indexSep = choice [char '-', char '.']
+indexSep = char '-' M.<|> char '.'
 
 form :: Parser Form
 form = maybeEmpty stringWSpaces
@@ -108,11 +111,11 @@ deprel = maybeEmpty deprel'
 
 deprel' :: Parser (Dep, Subtype)
 deprel' = do dep <- dep
-             st  <- option [] $ char ':' *> many1 letter
+             st  <- M.option [] $ char ':' *> M.some letterChar
              return (dep,st)
   where
     dep :: Parser Dep
-    dep = liftM mkDep $ many1 letter
+    dep = liftM mkDep $ M.some letterChar
 
 deps :: Parser Deps
 deps = listP $ listPair ':' index deprel'
@@ -124,30 +127,30 @@ misc = maybeEmpty stringWSpaces
 -- utility parsers
 litSpaces :: Parser ()
 -- because spaces consumes \t and \n
-litSpaces = skipMany $ char ' '
+litSpaces = M.skipMany $ char ' '
 
 commentPair :: Parser Comment
 commentPair =
-  keyValue '=' (stringNot "=\n\t") (option [] stringWSpaces)
+  keyValue '=' (stringNot "=\n\t") (M.option [] stringWSpaces)
 
 listPair :: Char -> Parser a -> Parser b -> Parser [(a, b)]
-listPair sep p q = sepBy1 (keyValue sep p q) (char '|')
+listPair sep p q = M.sepBy1 (keyValue sep p q) (char '|')
 
-stringNot :: String -> Parser String
+stringNot :: Stream -> Parser Stream
 -- [ ] second litSpaces in symbol is redundant
-stringNot s = symbol . many1 $ noneOf s
+stringNot s = symbol . M.some $ satisfy (\c -> not $ c `elem` s)
 
-stringWSpaces :: Parser String
+stringWSpaces :: Parser Stream
 stringWSpaces = stringNot "\t\n"
 
-stringWOSpaces :: Parser String
+stringWOSpaces :: Parser Stream
 stringWOSpaces = stringNot " \t\n"
 
 ---
 -- parser combinators
 keyValue :: Char -> Parser a -> Parser b -> Parser (a, b)
 keyValue sep p q = do key   <- p
-                      optional $ char sep
+                      M.optional $ char sep
                       value <- q
                       return (key, value)
 
@@ -155,7 +158,7 @@ symbol :: Parser a -> Parser a
 symbol p = litSpaces *> p <* litSpaces
 
 maybeEmpty :: Parser a -> Parser (Maybe a)
-maybeEmpty p = emptyField <|> liftM Just p
+maybeEmpty p = emptyField M.<|> liftM Just p
 
 listP :: Parser [a] -> Parser [a]
 -- using a parser that returns a possibly empty list like sepBy and
diff --git a/src/Conllu/Type.hs b/src/Conllu/Type.hs
index 75620ba..a4df8a3 100644
--- a/src/Conllu/Type.hs
+++ b/src/Conllu/Type.hs
@@ -1,3 +1,4 @@
+{-# LANGUAGE OverloadedStrings #-}
 module Conllu.Type where
 
 ---
@@ -10,11 +11,14 @@ import Data.List
 import Data.Maybe
 import Data.Ord
 import Data.Tree
+import qualified Data.Text as T
 
 ---
 -- type and data declarations
+type Stream = T.Text
+
 data Document = Document
-  { _file      :: String
+  { _file      :: Stream
   , _sents     :: [Sentence]
   } deriving (Eq,Show)
 
@@ -23,8 +27,8 @@ data Sentence = Sentence
   , _tokens :: [Token]
   } deriving (Eq, Show)
 
-type Comment    = StringPair
-type StringPair = (String, String)
+type Comment    = StreamPair
+type StreamPair = (Stream, Stream)
 
 data Token
   = SToken { _ix      :: Index
@@ -59,16 +63,16 @@ data Token
 
 type Index   = Int
 type IxSep   = Char
-type Form    = Maybe String
-type Lemma   = Maybe String
+type Form    = Maybe T.Text
+type Lemma   = Maybe T.Text
 type PosTag  = Maybe Pos
-type Xpostag = Maybe String
-type Feats   = [StringPair]
+type Xpostag = Maybe T.Text
+type Feats   = [StreamPair]
 type Dephead = Maybe Index
 type DepRel  = Maybe (Dep,Subtype)
-type Subtype = String
+type Subtype = T.Text
 type Deps    = [(Index,(Dep,Subtype))]
-type Misc    = Maybe String
+type Misc    = Maybe T.Text
 
 _dep :: Token -> Maybe Dep
 _dep = dep . _deprel
@@ -147,8 +151,8 @@ type ETree = (TTree, [Token]) -- enhanced tree
 
 ---
 -- constructor functions
-mkDep :: String -> Dep
-mkDep = mkDep' . upcaseString
+mkDep :: Stream -> Dep
+mkDep = mkDep' . upcaseStream
   where
     mkDep' "ACL"        = ACL
     mkDep' "ADVCL"      = ADVCL
@@ -188,8 +192,8 @@ mkDep = mkDep' . upcaseString
     mkDep' "VOCATIVE"   = VOCATIVE
     mkDep' "XCOMP"      = XCOMP
 
-mkPos :: String -> Pos
-mkPos = mkPos' . upcaseString
+mkPos :: Stream -> Pos
+mkPos = mkPos' . upcaseStream
   where
     mkPos' "ADJ"      = ADJ
     mkPos' "ADP"      = ADP
@@ -270,8 +274,8 @@ eTokenOK h dr d = assNothing h $ assNothing dr $ assSomething d True
 
 ---
 -- utility functions
-upcaseString :: String -> String
-upcaseString = map toUpper
+upcaseStream :: Stream -> Stream
+upcaseStream = T.toUpper
 
 assNothing :: Maybe a -> Bool -> Bool
 assNothing m = assert (isNothing m)