EDL to CSV
via Edit decision list (report only)- Kept
- Frame-accurate timing, Clip boundaries, Clip names, Media paths / reels, Transitions
- Changed
- nothing
What survives each conversion
Every CutConvert pair, with what the input carries, what the output keeps, and what changes on the way. Read it before you convert a broadcast delivery or a transcript with speakers, or trust the same line under the preview on any converter page.
44 converters · 28 formats · 15 reported changes
Formats and what they can hold
| Format | Hub | ms | frames | words | speaker | position | style | breaks | clips | names | paths | trans. | fx | markers | tracks |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| SRT | timed text | ● | · | · | · | · | ● | ● | · | · | · | · | · | · | · |
| WebVTT | timed text | ● | · | · | ● | ● | ● | ● | · | · | · | · | · | · | · |
| YouTube SBV | timed text | ● | · | · | · | · | · | ● | · | · | · | · | · | · | · |
| ASS / SSA | timed text | ● | · | · | ● | ● | ● | ● | · | · | · | · | · | · | · |
| SCC (CEA-608) | timed text | · | ● | · | · | ● | ● | ● | · | · | · | · | · | · | ● |
| MCC (CEA-708) | timed text | · | ● | · | · | ● | ● | ● | · | · | · | · | · | · | ● |
| EBU-STL | timed text | · | ● | · | · | ● | ● | ● | · | · | · | · | · | · | · |
| iTT | timed text | · | ● | · | · | ● | ● | ● | · | · | · | · | · | · | · |
| TTML | timed text | ● | · | · | ● | ● | ● | ● | · | · | · | · | · | · | · |
| Timed CSV | timed text | ● | · | · | ● | · | · | ● | · | · | · | · | · | · | · |
| CSV | timed text | ● | · | · | ● | · | · | ● | · | · | · | · | · | · | · |
| Rev transcript TXT | timed text | ● | · | · | ● | · | · | · | · | · | · | · | · | · | · |
| Otter transcript TXT | timed text | ● | · | · | ● | · | · | · | · | · | · | · | · | · | · |
| Word-style transcript | timed text | ● | · | · | ● | · | · | · | · | · | · | · | · | · | · |
| Plain text | timed text | · | · | · | ● | · | · | ● | · | · | · | · | · | · | · |
| Word DOCX | timed text | ● | · | · | ● | · | · | ● | · | · | · | · | · | · | · |
| Premiere transcript JSON | timed text | ● | · | ● | ● | · | · | · | · | · | · | · | · | · | · |
| Premiere transcript JSON | timed text | ● | · | ● | ● | · | · | · | · | · | · | · | · | · | · |
| Premiere .prtranscript | timed text | ● | · | ● | ● | · | · | · | · | · | · | · | · | · | · |
| Transcript JSON (Whisper, AssemblyAI, …) | timed text | ● | · | ● | ● | · | · | · | · | · | · | · | · | · | · |
| Avid caption TXT | timed text | · | ● | · | · | · | · | ● | · | · | · | · | · | · | · |
| FCPXML (titles) | timed text | · | ● | · | · | ● | ● | ● | · | · | · | · | · | · | · |
| Premiere project | timeline | · | ● | · | · | · | · | · | ● | ● | ● | ● | ● | ● | ● |
| Avid AAF | timeline | · | ● | · | · | · | · | · | ● | ● | ● | ● | ● | ● | ● |
| Resolve project | timeline | · | ● | · | · | · | · | · | ● | ● | ● | ● | ● | ● | ● |
| FCP7 XML (xmeml) | timeline | · | ● | · | · | · | · | · | ● | ● | ● | ● | · | ● | ● |
| EDL (CMX3600) | EDL | · | ● | · | · | · | · | · | ● | ● | ● | ● | · | · | · |
| EDL event CSV | EDL | · | ● | · | · | · | · | · | ● | ● | ● | ● | · | · | · |
Every pair
TXT to SRT
via Timed text (cue model)- Kept
- Millisecond timing
- Changed
- Speaker names written into the text as a NAME: prefix, not a separate field
- Filled in
- Line breaks inside a cue (the output has a place for it; the source did not)
- Kept
- Millisecond timing
- Changed
- Speaker names written into the text as a NAME: prefix, not a separate field
- Filled in
- Line breaks inside a cue (the output has a place for it; the source did not)
- Kept
- Millisecond timing
- Changed
- Speaker names written into the text as a NAME: prefix, not a separate field
- Filled in
- Line breaks inside a cue (the output has a place for it; the source did not)
TXT to Premiere JSON
via Timed text (cue model)Rev transcript TXT → Premiere transcript JSON
- Kept
- Millisecond timing, Speaker names
- Changed
- nothing
Otter transcript TXT → Premiere transcript JSON
- Kept
- Millisecond timing, Speaker names
- Changed
- nothing
Word-style transcript → Premiere transcript JSON
- Kept
- Millisecond timing, Speaker names
- Changed
- nothing
DOC to Premiere JSON
via Timed text (cue model)Word-style transcript → Premiere transcript JSON
- Kept
- Millisecond timing, Speaker names
- Changed
- nothing
SRT to Avid TXT
via Timed text (cue model)- Kept
- Line breaks inside a cue
- Changed
- Millisecond timing timing is snapped to frames at the chosen or detected rate reported
- Styling (italics, colour) inline tags are stripped
- Filled in
- Frame-accurate timing (the output has a place for it; the source did not)
SRT to Premiere JSON
via Timed text (cue model)SRT → Premiere transcript JSON
- Kept
- Millisecond timing
- Changed
- Styling (italics, colour) inline tags are stripped
- Line breaks inside a cue Premiere transcript JSON has no place for it
- Filled in
- Speaker names (the output has a place for it; the source did not)
SRT to VTT
via Timed text (cue model)- Kept
- Millisecond timing, Line breaks inside a cue
- Changed
- Styling (italics, colour) inline tags are stripped
SRT to Text
via Timed text (cue model)- Kept
- Line breaks inside a cue
- Changed
- Millisecond timing plain text has no timing
- Styling (italics, colour) inline tags are stripped
- Filled in
- Speaker names (the output has a place for it; the source did not)
SBV to SRT
via Timed text (cue model)- Kept
- Millisecond timing, Line breaks inside a cue
- Changed
- nothing
VTT to SRT
via Timed text (cue model)- Kept
- Millisecond timing, Line breaks inside a cue
- Changed
- Speaker names voice tags are stripped with the other inline tags
- On-screen position placement is not carried
- Styling (italics, colour) inline tags are stripped
VTT to Text
via Timed text (cue model)- Kept
- Line breaks inside a cue
- Changed
- Millisecond timing plain text has no timing
- Speaker names voice tags are stripped with the other inline tags
- On-screen position placement is not carried
- Styling (italics, colour) inline tags are stripped
Premiere Transcript to SRT
via Timed text (cue model)- Kept
- Millisecond timing
- Changed
- Per-word timestamps words are grouped into cues; per-word times are not kept
- Speaker names written into the text as a NAME: prefix, not a separate field
- Filled in
- Line breaks inside a cue (the output has a place for it; the source did not)
Premiere transcript JSON → SRT
- Kept
- Millisecond timing
- Changed
- Per-word timestamps words are grouped into cues; per-word times are not kept
- Speaker names written into the text as a NAME: prefix, not a separate field
- Filled in
- Line breaks inside a cue (the output has a place for it; the source did not)
Premiere Transcript to VTT
via Timed text (cue model)Premiere .prtranscript → WebVTT
- Kept
- Millisecond timing
- Changed
- Per-word timestamps words are grouped into cues; per-word times are not kept
- Speaker names written into the text as a NAME: prefix, not a separate field
- Filled in
- Line breaks inside a cue (the output has a place for it; the source did not)
Premiere transcript JSON → WebVTT
- Kept
- Millisecond timing
- Changed
- Per-word timestamps words are grouped into cues; per-word times are not kept
- Speaker names written into the text as a NAME: prefix, not a separate field
- Filled in
- Line breaks inside a cue (the output has a place for it; the source did not)
Premiere Transcript to Text
via Timed text (cue model)Premiere .prtranscript → Plain text
- Kept
- Speaker names
- Changed
- Millisecond timing plain text has no timing
- Per-word timestamps words are grouped into cues; per-word times are not kept
- Filled in
- Line breaks inside a cue (the output has a place for it; the source did not)
Premiere transcript JSON → Plain text
- Kept
- Speaker names
- Changed
- Millisecond timing plain text has no timing
- Per-word timestamps words are grouped into cues; per-word times are not kept
- Filled in
- Line breaks inside a cue (the output has a place for it; the source did not)
Premiere Transcript to Word
via Timed text (cue model)Premiere .prtranscript → Word DOCX
- Kept
- Millisecond timing, Speaker names
- Changed
- Per-word timestamps words are grouped into cues; per-word times are not kept
- Filled in
- Line breaks inside a cue (the output has a place for it; the source did not)
Premiere transcript JSON → Word DOCX
- Kept
- Millisecond timing, Speaker names
- Changed
- Per-word timestamps words are grouped into cues; per-word times are not kept
- Filled in
- Line breaks inside a cue (the output has a place for it; the source did not)
Premiere Transcript to CSV
via Timed text (cue model)- Kept
- Millisecond timing, Speaker names
- Changed
- Per-word timestamps words are grouped into cues; per-word times are not kept
- Filled in
- Line breaks inside a cue (the output has a place for it; the source did not)
Premiere transcript JSON → CSV
- Kept
- Millisecond timing, Speaker names
- Changed
- Per-word timestamps words are grouped into cues; per-word times are not kept
- Filled in
- Line breaks inside a cue (the output has a place for it; the source did not)
SRT to Word
via Timed text (cue model)- Kept
- Millisecond timing, Line breaks inside a cue
- Changed
- Styling (italics, colour) inline tags are stripped
- Filled in
- Speaker names (the output has a place for it; the source did not)
VTT to Word
via Timed text (cue model)- Kept
- Millisecond timing, Line breaks inside a cue
- Changed
- Speaker names voice tags are stripped with the other inline tags
- On-screen position placement is not carried
- Styling (italics, colour) inline tags are stripped
ASS to SRT
via Timed text (cue model)- Kept
- Millisecond timing, Line breaks inside a cue
- Changed
- Speaker names written into the text as a NAME: prefix, not a separate field
- On-screen position placement is not carried
- Styling (italics, colour) override tags are stripped
JSON to SRT
via Timed text (cue model)Transcript JSON (Whisper, AssemblyAI, …) → SRT
- Kept
- Millisecond timing
- Changed
- Per-word timestamps words are grouped into cues; per-word times are not kept
- Speaker names written into the text as a NAME: prefix, not a separate field
- Filled in
- Line breaks inside a cue (the output has a place for it; the source did not)
JSON to VTT
via Timed text (cue model)Transcript JSON (Whisper, AssemblyAI, …) → WebVTT
- Kept
- Millisecond timing
- Changed
- Per-word timestamps words are grouped into cues; per-word times are not kept
- Speaker names written into the text as a NAME: prefix, not a separate field
- Filled in
- Line breaks inside a cue (the output has a place for it; the source did not)
JSON to Word
via Timed text (cue model)Transcript JSON (Whisper, AssemblyAI, …) → Word DOCX
- Kept
- Millisecond timing, Speaker names
- Changed
- Per-word timestamps words are grouped into cues; per-word times are not kept
- Filled in
- Line breaks inside a cue (the output has a place for it; the source did not)
JSON to Text
via Timed text (cue model)Transcript JSON (Whisper, AssemblyAI, …) → Plain text
- Kept
- Speaker names
- Changed
- Millisecond timing plain text has no timing
- Per-word timestamps words are grouped into cues; per-word times are not kept
- Filled in
- Line breaks inside a cue (the output has a place for it; the source did not)
CapCut to SRT
via Timed text (cue model)Transcript JSON (Whisper, AssemblyAI, …) → SRT
- Kept
- Millisecond timing
- Changed
- Per-word timestamps words are grouped into cues; per-word times are not kept
- Speaker names written into the text as a NAME: prefix, not a separate field
- Filled in
- Line breaks inside a cue (the output has a place for it; the source did not)
SRT to CSV
via Timed text (cue model)- Kept
- Millisecond timing, Line breaks inside a cue
- Changed
- Styling (italics, colour) inline tags are stripped
- Filled in
- Speaker names (the output has a place for it; the source did not)
CSV to SRT
via Timed text (cue model)- Kept
- Millisecond timing, Line breaks inside a cue
- Changed
- Speaker names written into the text as a NAME: prefix, not a separate field
SRT to FCPXML
via Timed text (cue model)- Kept
- Line breaks inside a cue
- Changed
- Millisecond timing timing is snapped to frames reported
- Styling (italics, colour) inline tags are stripped
- Filled in
- Frame-accurate timing (the output has a place for it; the source did not)
FCPXML to SRT
via Timed text (cue model)- Kept
- Line breaks inside a cue
- Changed
- Frame-accurate timing SRT has no place for it
- On-screen position placement is not carried
- Styling (italics, colour) inline tags are stripped
- Filled in
- Millisecond timing (the output has a place for it; the source did not)
FCPXML to Word
via Timed text (cue model)- Kept
- Line breaks inside a cue
- Changed
- Frame-accurate timing Word DOCX has no place for it
- On-screen position placement is not carried
- Styling (italics, colour) inline tags are stripped
- Filled in
- Millisecond timing, Speaker names (the output has a place for it; the source did not)
FCPXML to CSV
via Timed text (cue model)- Kept
- Line breaks inside a cue
- Changed
- Frame-accurate timing CSV has no place for it
- On-screen position placement is not carried
- Styling (italics, colour) inline tags are stripped
- Filled in
- Millisecond timing, Speaker names (the output has a place for it; the source did not)
SRT to SCC
via Timed text (cue model)- Kept
- nothing the source had survives unchanged
- Changed
- Millisecond timing snapped to 29.97 drop-frame; a caption can be delayed to fit the 608 data rate reported
- Styling (italics, colour) inline tags are stripped
- Line breaks inside a cue wrapped to 32 columns; long cues are split reported
- Filled in
- Frame-accurate timing (the output has a place for it; the source did not)
SCC to SRT
via Timed text (cue model)- Kept
- Line breaks inside a cue
- Changed
- Frame-accurate timing SRT has no place for it
- On-screen position placement is not carried
- Styling (italics, colour) inline tags are stripped
- Multiple tracks / channels only CC1 is decoded; a second channel is ignored reported
- Filled in
- Millisecond timing (the output has a place for it; the source did not)
SCC to CSV
via Timed text (cue model)- Kept
- Line breaks inside a cue
- Changed
- Frame-accurate timing CSV has no place for it
- On-screen position placement is not carried
- Styling (italics, colour) inline tags are stripped
- Multiple tracks / channels only CC1 is decoded; a second channel is ignored reported
- Filled in
- Millisecond timing, Speaker names (the output has a place for it; the source did not)
SRT to MCC
via Timed text (cue model)- Kept
- nothing the source had survives unchanged
- Changed
- Millisecond timing snapped to 30 drop-frame; a caption can be delayed to fit the 608 data rate reported
- Styling (italics, colour) inline tags are stripped
- Line breaks inside a cue wrapped to 32 columns; long cues are split reported
- Filled in
- Frame-accurate timing (the output has a place for it; the source did not)
MCC to SRT
via Timed text (cue model)- Kept
- Line breaks inside a cue
- Changed
- Frame-accurate timing SRT has no place for it
- On-screen position placement is not carried
- Styling (italics, colour) inline tags are stripped
- Multiple tracks / channels the 608 compatibility track is decoded; native 708 services are not reported
- Filled in
- Millisecond timing (the output has a place for it; the source did not)
MCC to CSV
via Timed text (cue model)- Kept
- Line breaks inside a cue
- Changed
- Frame-accurate timing CSV has no place for it
- On-screen position placement is not carried
- Styling (italics, colour) inline tags are stripped
- Multiple tracks / channels the 608 compatibility track is decoded; native 708 services are not reported
- Filled in
- Millisecond timing, Speaker names (the output has a place for it; the source did not)
SRT to STL
via Timed text (cue model)- Kept
- nothing the source had survives unchanged
- Changed
- Millisecond timing snapped to 25 fps
- Styling (italics, colour) inline tags are stripped
- Line breaks inside a cue wrapped to 40 characters over 4 rows; long cues are split reported
- Filled in
- Frame-accurate timing (the output has a place for it; the source did not)
STL to SRT
via Timed text (cue model)- Kept
- Line breaks inside a cue
- Changed
- Frame-accurate timing SRT has no place for it
- On-screen position placement is not carried
- Styling (italics, colour) inline tags are stripped
- Filled in
- Millisecond timing (the output has a place for it; the source did not)
STL to CSV
via Timed text (cue model)- Kept
- Line breaks inside a cue
- Changed
- Frame-accurate timing CSV has no place for it
- On-screen position placement is not carried
- Styling (italics, colour) inline tags are stripped
- Filled in
- Millisecond timing, Speaker names (the output has a place for it; the source did not)
Premiere Project to XML
via Timeline (clips on tracks)Premiere project → FCP7 XML (xmeml)
- Kept
- Frame-accurate timing, Clip boundaries, Clip names, Multiple tracks / channels
- Changed
- Media paths / reels synthetic or offline media has no file path reported
- Transitions cuts only: a transition becomes a hard cut
- Effects and retimes effects, retimes, and nested sequences are not carried
- Markers markers are not read
Premiere Project to DRP
via Timeline (clips on tracks)Premiere project → Resolve project
- Kept
- Frame-accurate timing, Clip boundaries, Clip names, Multiple tracks / channels
- Changed
- Media paths / reels synthetic or offline media has no file path reported
- Transitions cuts only: a transition becomes a hard cut
- Effects and retimes effects, retimes, and nested sequences are not carried
- Markers markers are not read
DaVinci Resolve Project to XML
via Timeline (clips on tracks)Resolve project → FCP7 XML (xmeml)
- Kept
- Clip boundaries, Clip names, Media paths / reels
- Changed
- Frame-accurate timing timelines at a different frame rate from the first are left out reported
- Transitions cuts only
- Effects and retimes effects and grades are not carried
- Markers markers are not read
- Multiple tracks / channels subtitle tracks are left out; video and audio tracks are carried reported
AAF to XML
via Timeline (clips on tracks)- Kept
- Frame-accurate timing, Clip boundaries, Clip names, Media paths / reels, Multiple tracks / channels
- Changed
- Transitions a transition becomes a straight cut at its midpoint reported
- Effects and retimes an effect is flattened to its first underlying clip reported
- Markers markers are not read
AAF to DRP
via Timeline (clips on tracks)- Kept
- Frame-accurate timing, Clip boundaries, Clip names, Media paths / reels, Multiple tracks / channels
- Changed
- Transitions a transition becomes a straight cut at its midpoint reported
- Effects and retimes an effect is flattened to its first underlying clip reported
- Markers markers are not read
Refused pairs
Any EDL to any timed-text format, and any timed-text format to EDL. An EDL is a cut list and captions are timed text; a file that opens but means nothing is worse than a refusal. The timeline hub (Premiere, AAF, Resolve, FCP7 XML) likewise never converts to or from timed text; the only bridge is titles inside FCPXML, which travel as cues.
How to read the table
Every format belongs to one of three hubs. Timed text is words between two timestamps: SRT, WebVTT, SCC, transcripts, DOCX and the rest. Timeline is a sequence of clips with media paths, transitions and markers: Premiere projects, Avid AAF, Resolve projects, FCP7 XML. EDL is a cut list. A conversion reads the input into its hub, then writes the output from that hub, and the table shows what the input had that the output can still hold.
Hubs never cross. An EDL never becomes an SRT and an SRT never becomes an EDL, in either direction, because neither carries what the other needs and a file that opens but means nothing is worse than a refusal. The only bridge is titles inside an FCPXML, which travel as timed text. Timelines likewise never convert to or from captions.
The first table lists what each format can hold. The cards below it list every pair, with the capabilities kept, the ones changed and why, and whether the change is reported on the result. The same line appears under the live preview on every converter page, before anything is converted. Files never leave your browser.
FAQ
Why can't EDL convert to SRT?
An EDL is a cut list: reel names, source and record timecodes, transitions. An SRT is timed text: words on screen between two timestamps. Neither carries what the other needs, so a file that "converted" would open and mean nothing. CutConvert refuses the pair in both directions and sends EDLs to the EDL to CSV report instead.
Which conversions keep speaker names?
Any pair where both formats have a place for a speaker: transcript JSON, Premiere transcripts, Rev, Otter and Word-style transcripts, WebVTT, ASS, TTML, timed CSV and Word DOCX. Writing to SRT, SBV, SCC, MCC, EBU-STL, iTT or Avid caption TXT folds the name into the text as a NAME: prefix or drops it; the row for that pair says which.
Which conversions keep styling?
Italics, bold and colour survive between SRT, WebVTT, ASS, TTML, SCC, MCC, EBU-STL, iTT and FCPXML titles, within what each format can express. Anything written to plain text, CSV, DOCX, Avid caption TXT or a transcript JSON loses inline styling, and the table marks that change on the pair.
What does "reported" mean?
A change tagged "reported" is announced on the result the moment it happens: the converted file carries a warning naming what changed. A change without the tag is still shown here and in the preview line under the file before you convert, but the finished file will not carry a note about it.
Is this table hand-written?
No. It is generated from the same capability table the converters and the preview line read, so a pair here matches what the converter actually does. Files are converted in your browser and never uploaded.
Related tools & converters
- SRT convertersEvery converter that reads or writes SRT.
- EDL toolsEDL to CSV reports and the EDL guides.
- Premiere transcriptsTurn .prtranscript and transcript JSON into captions and documents.
- SRT to SCCBroadcast CEA-608 captions from an SRT, with the changes named.
- JSON to SRTWhisper, AssemblyAI and Deepgram transcripts into subtitles.
- GuidesStep-by-step workflows for each pair.