Subtitle text encoding defines how stored bytes are interpreted as written characters. UTF-8 is common, but older files may use other encodings that a player cannot reliably infer. Interpreting the wrong encoding can produce garbled accents, unexpected symbols or replacement characters even when cue timing is correct. This is separate from the subtitle file format: two SRT files can use different character encodings. It also differs from font support, where the text may be decoded correctly but the selected font lacks the glyphs needed to draw particular characters.
Subtitle encoding and Font
Encoding determines character identity; a font determines how supported characters are drawn.
Bytes, characters and glyphs
A parser reads bytes, decodes them into characters and then asks a renderer to draw those characters. A problem at the decoding stage often turns recognizable words into sequences of wrong symbols. A rendering problem may instead show empty boxes for particular scripts. Distinguishing the stages helps avoid repeatedly converting a valid file when the missing component is a suitable font.
Renaming does not change the text
Changing .srt to another extension does not transcode character bytes. Likewise, saving garbled text after opening it with the wrong interpretation can permanently preserve the wrong characters. Work from the original file and choose the correct source encoding before converting to UTF-8 where appropriate. Keep a backup so an unsuccessful conversion does not replace the only intact version.
Use representative characters
Inspect lines containing accents, punctuation and the relevant writing system rather than an ASCII-only opening. If ordinary letters work but every accented character fails, encoding is a useful suspect. If the whole track is absent, check parsing and source selection first. After conversion, reopen the result independently and confirm both the wording and cue timing remain intact.
An illustrative example
An SRT file displays plain English words correctly but turns accented names into strange symbol pairs. Its timestamps still work. Opening the original with the correct character encoding and exporting valid UTF-8 repairs the names; changing subtitle delay would never affect the character interpretation problem.
Practical actions
- Inspect accented and non-Latin characters in the original file.
- Distinguish wrong characters from missing-font boxes.
- Back up the original before transcoding to UTF-8.