Build an accurate speech to text analyzer

I am building an application in C# that requires me to analyze audio fragments and get the spoken words from it. In my test case I have 5 audio fragments (.wav) extracted from a movie that each hold 20 seconds of audio. I need to have a module that takes these audio fragments and performs a speecht to text analysis on it. I need to have a listing of all sentences from the fragments each with start time and duration. I already tried this with the builtin C# module [login to view URL], but got results that were too inaccurate. You can use any kind of library or module you want, the most important thing is to have a quite decent accuracy. My audio fragments I test with can be found here: [login to view URL]!dpxWDAzL!4vRoa53qxh-6_oFShmXj0l_QqA6why7ndcAZZb7g1ZE [login to view URL]!thBiQB4Q!duh7myW-_yyS4Ht3Q0DXfgsSICvXpidac12o_p-ejI0 [login to view URL]!80QQHTQL!jxJRRA82FUOmUsR0HmuG2umKRAsIQ-o0m1iB_bdLWYA