SELECT Regexp_substr('I know English very well.', '[[:alpha:]]+(gli)[[:alpha:]]+' 
       , 1, 1, 
              'i') "REGEXP_INSTR" 
FROM   dual;

  WITH string_tab 
     AS (SELECT 1                                   id, 
                'I watched English Winglish.' val 
         FROM   dual 
         UNION 
         SELECT 2                           ID, 
                'I know English very well.' val 
         FROM   dual), 
     --seaching for the given string in the sentence 
     searched_string 
     AS (SELECT id, 
                LEVEL 
                   occurance, 
                val, 
                Regexp_substr(val, '[[:alpha:]]*(gli)[[:alpha:]]*', 1, LEVEL, 
                'i') 
                VALUE 
         FROM   string_tab 
         CONNECT BY LEVEL < Regexp_count(val, '(gli)') + 1 
                    AND PRIOR id = id 
                              AND PRIOR dbms_random.value IS NOT NULL) 
--grouping the searched words 
SELECT id, 
       val, 
       Listagg (value, ',') 
         within GROUP ( ORDER BY occurance ) Words
FROM   searched_string 
GROUP  BY id, 
          val;