Subtitle Converter

SRT is the safest choice for almost everything: VLC, PotPlayer, KMPlayer, Plex, most smart TVs and video editors all read it, and it is what most subtitle sites offer. SMI (SAMI) is the format many Korean subtitle makers used for years, and some Korean players and older TVs still expect it, so convert SRT to SMI when a device ignores your SRT. WebVTT (.vtt) is the format web players and HTML5 video use, so it is the one to pick for a website or a course platform. ASS and SSA come from anime fansubbing and support positioning and karaoke effects; this tool reads the text and timing from them, while the positioning stays behind because other formats cannot store it. SBV is what YouTube Studio exports and accepts, and LRC is the lyrics format music players use.

When you convert to a simpler format, the words and timings carry over exactly, while italics, bold and colours carry over only where the target supports them. SRT and SMI keep all three, VTT keeps italics and bold, ASS turns them into its own tags, and SBV, LRC and plain text drop them.

If every line appears the same amount too early or too late, the subtitle was made for a release with a different intro or ad break, and a single shift fixes it. Watch the first spoken line, note how far off it is, and enter that number of seconds: a positive number delays the subtitles and a negative number brings them forward. The preview table shows the new times straight away, so you can compare them with the video before downloading.

If the first lines match but the gap grows to several seconds by the end, the subtitle was timed for a different frame rate, usually 25 fps (European TV) against 23.976 fps (film and most streaming releases). Choose the subtitle's frame rate and the video's frame rate, and every time is stretched by the right ratio. You can combine both fixes, and they apply to every file in the batch.

Older subtitle files were often saved in a regional encoding instead of UTF-8: Windows-1252 for Western European languages, EUC-KR for Korean, Shift_JIS for Japanese. Opened on a Mac, a phone or a newer TV, those files turn accents into strings like é and other alphabets into rows of symbols. The converter detects the encoding of each file before reading it, so the text comes out correctly, and the file list shows which encoding it found. The new file is saved as UTF-8 with a BOM for SRT, SMI and ASS, which Windows players and TVs recognise, and as plain UTF-8 for the other formats. An EUC-KR option is there for older Korean TVs and players that cannot read UTF-8; emoji and some rare characters cannot be stored in it, and the tool tells you how many were replaced.

SMI files can also hold two languages at once, typically Korean (KRCC) and English (ENCC). When a file has more than one, a language choice appears, and lines that the subtitle maker switched off inside comments are skipped rather than shown.

We ran a full-length French film subtitle with 1,332 lines through every format and back again. In SRT, VTT, SMI and SBV every start and end time came back identical to the millisecond, and in ASS and LRC, which store hundredths of a second, every time stayed within 10 milliseconds. The same file saved in Windows-1252 and in UTF-16 was read with all its accented letters intact. A Korean SMI saved in EUC-KR came out with correct Hangul, and a two-language SMI offered Korean and English separately while skipping a line hidden in a comment. The one limit we found is LRC: lyrics files have no end times, so lines that overlap in the original cannot both be kept.