Scribis Subtitle Review and Export: From Transcript Edits to SRT
Use the Scribis Translation editor to replay audio, correct subtitle text and timing, review speakers, and export subtitles or a video with burned-in captions.
Recognition is only the first step in making subtitles. Scribis's Translation page combines a video preview, audio waveform, and subtitle list so you can listen to the source while editing text and timing. These steps are for a completed transcription. If the task is still running or has not produced subtitles, resolve that first.
1. Start with playback
Open the transcription result and play a short section to confirm that the video, audio, and subtitle timeline are in sync. Use the subtitle list to locate a sentence, and the waveform and player to check when it is actually spoken. On the first pass, note obvious issues instead of editing every word immediately; keeping the full context helps you make better decisions.
Check names, product terms, numbers, negations, and mixed-language terms first. A transcript can sound natural and still contain errors. Replay a few seconds before and after a passage when you need more context.
2. Correct text and subtitle timing
Select a subtitle row to edit its text or adjust its start and end times. If a waveform or timeline editor is available, locate the beginning of the speech and extend or shorten the subtitle as needed. Avoid showing it too early or leaving it on screen after the next speaker has started.
Try to keep one complete idea in each subtitle line. Split long sentences at natural pauses; if short subtitles flash by too quickly, consider merging adjacent segments. Replay the passage after editing to make sure the reading time matches the speech. Use undo and redo to compare changes, and avoid overwriting the source file if you are unsure.
3. Review speakers and translations
For conversations with multiple speakers, check the labels throughout, especially during turn-taking, interruptions, and background speech. Diarization provides a clue about speaker changes; it does not verify anyone's identity. Use the recording and the context you know to confirm the labels.
If you have translated subtitles, check that proper nouns and numbers match, then review tone, negations, and references across sentences. Translation or batch tools can speed up a first draft for longer material, but review the context before publishing rather than checking each sentence in isolation.
4. Choose subtitle styling and an export format
For YouTube or video editing software, export a sidecar subtitle format such as SRT or VTT. To keep subtitles permanently visible in the finished video, export a video with burned-in captions; unlike sidecar subtitles, these cannot be turned off or edited separately afterward. Different formats and video exports may require different system components, so confirm the available options and save location before exporting.
Reopen the exported subtitle file and check the text and timestamps near the beginning, middle, and end. For bilingual subtitles, make sure the languages do not overlap and check sentence order and line length. Keep the original recognition result and save the reviewed version separately so you can compare or revise it later.
Quick review checklist
- Do names, specialized terms, and numbers match the audio?
- Do subtitle start and end times match the speech?
- Are the line breaks natural and easy to read?
- Have speaker labels and translations been checked by a person?
- Did you reopen and review the exported subtitle file?
The goal of subtitle editing is not to put every pause on screen. It is to help viewers follow the content while keeping text, audio, and video in sync.