Turkish subtitles use ISO-8859-9
Turkish subtitles, found on internet, are generally encoded using ISO-8859-9.
But it seems like trying to decode subtitles as UTF-8. The root cause should be my UTF-8 environment. This cause some Turkish characters (like "ş", "ğ", "ı") wrongly interpreted.
There were two similar issues about Greek (#3860 (closed), ISO-8859-7) and Hebrew (#7474 (closed), ISO-8859-8) subtitles.
Bekir Dogan