Premiere Pro says "Unsupported format or damaged file" when you import captions
“Unsupported format or damaged file”
What is actually wrong
Premiere Pro only imports a short list of caption formats: SRT, SCC, MCC, EBU STL, DFXP/TTML and its own caption XML. A .vtt, .ass, .sbv, .json or .txt file is not damaged; it is simply a format the caption importer never looks inside. The message is the same one Premiere shows for any media it cannot decode, which is why it looks like corruption.
Fix it here: VTT to SRT
Drop the file. It is detected and converted in your browser; nothing is uploaded. Free to start, no signup.
Convert a file now
Conversions …
Drop EDL, SRT, VTT, .prtranscript, or transcript files
Any caption, transcript or timeline file: CutConvert picks the converter and shows what survives · up to 100 KB per file
The fix, step by step
Check the extension
Look at the file Premiere rejected. If it ends in .vtt, .ass, .ssa, .sbv, .json or .txt, the format is the problem, not the file.
Convert it to SRT
Drop the file into the converter on this page. It detects the format and writes a standard SRT with the same timing. Nothing is uploaded.
Import the SRT
In Premiere, choose File > Import (or drag the SRT into the Project panel), then drag the caption item onto the timeline above the sequence.
What Premiere's caption importer accepts
Adobe documents the caption formats Premiere Pro can import: SubRip (.srt), Scenarist (.scc), MacCaption (.mcc), EBU STL (.stl), DFXP/TTML (.dfxp, .xml) and Premiere's own caption XML. Everything else falls through to the generic media importer, which does not understand text files and reports them as unsupported or damaged.
The formats that trigger this most often in a post house:
- WebVTT (.vtt) from web players, Zoom, YouTube or a transcription API. It looks almost identical to SRT, but the header line, the dots in the timestamps and the optional cue settings are enough to break the import.
- ASS/SSA (.ass, .ssa) from Aegisub or a fansub-style workflow. The timing lives inside
Dialogue:lines with override tags Premiere has never heard of. - SBV (.sbv) downloaded from YouTube Studio. Same idea as SRT, different timestamp syntax.
- JSON from Whisper, AssemblyAI, Deepgram or Rev.ai. Editors rename it to .srt hoping Premiere will read it; the importer opens it, finds braces instead of cue numbers, and gives up.
Why the file is not actually damaged
A caption file that opens in a text editor and shows readable cues is fine. Premiere's message is generic: it is emitted whenever the import pipeline finds no decoder for the file, whether that is a truncated video or a perfectly valid VTT. Treat it as "unknown format", not "corrupt".
There is one real damage case worth ruling out: a file saved by Word or a rich-text editor, which wraps the cues in RTF or DOCX markup. Open it in a plain text editor. If you see cue numbers and --> arrows, convert it. If you see formatting codes, save it as plain text first.
The fix
Convert the file to SRT. The converter on this page reads VTT, ASS/SSA, SBV and transcript JSON, keeps the original timing to the millisecond, and strips styling Premiere would not display anyway. The conversion runs entirely in your browser, so the file never leaves your machine, which matters when the captions are for an unreleased cut.
Once you have the SRT, import it through File > Import, drop it onto the timeline, and Premiere creates a caption track from it. If Premiere then complains about the frame rate, that is a separate problem: caption files carry time, not frames, and Premiere maps them to your sequence when you place them.
FAQ
- Does Premiere Pro import VTT files?
- No. WebVTT is not in the caption import list. Convert it to SRT first, which keeps the timing and text exactly.
- Can I just rename .vtt or .json to .srt?
- Renaming does not change the contents. Premiere will open the file, fail to parse it and show the same error, or import an empty caption track.
- Is the converted SRT uploaded anywhere?
- No. Detection and conversion run in your browser. The site only records that a conversion happened, never the file.