Upload your recording.
MP3, WAV, M4A, MP4, WebM, OGG, MPEG, or MPGA, up to 50 MB and 20 minutes. Video files are read for their audio track only.
Turn audio into subtitles you can edit.
Drop in a recording, pick the spoken language, and Cue writes a timed subtitle draft. Check the names and numbers, fix what needs fixing, and export a clean SRT, VTT, or TXT file. Already have subtitles and only need them translated? Start with the subtitle translator instead.
THE IMPORTANT PART
A draft you can work with
Cue’s transcription is AI output, so it is a strong first draft and still a draft. The review step sits in the middle of the workflow on purpose.
HOW IT WORKS
Four steps to a file you can use.
Pick the spoken language.
Pick the spoken language when you know it, or let Cue try to detect it. We recommend selecting the language you know rather than leaving it to detection.
Add the details.
Names, places, and unusual terms go in the glossary field. Ask for a translation in the same step if you need one.
Review and export.
Click each line to check it, fix the words or the timing, and export SRT, VTT, or TXT. Exports are free and carry no watermark.
THE DETAILS
Questions worth answering.
What can I upload?
MP3, WAV, M4A, MP4, WebM, OGG, MPEG, and MPGA files, up to 50 MB and 20 minutes. MP4 and WebM files are read for their audio track; Cue does not render or edit video.
What if I do not know the spoken language?
Use Detect automatically, or select the language when you know it. We recommend selecting it. In one of our own tests, a Spanish recording with automatic detection came back labeled English; that is a single observed case, not a measured accuracy comparison.
How accurate is the transcription?
It varies by recording. In our own small test set, a clean English recording dropped two sentences, while an accented, noisier recording stayed close to the source. That test was a handful of synthetic recordings, not a benchmark for your audio. That is why the workflow runs in this order: generate the draft, review and edit it, then export.
Can I use it for interviews and podcasts?
Yes, when speech is the main content. Two things to know: Cue processes one audio track, and recordings longer than 20 minutes need splitting first.
Do you burn subtitles into my video?
No. Cue works on subtitle files. You get SRT, VTT, or TXT and pair them with your video in your own editor.
Can I translate the result?
Yes. Choose a target language when you upload (2 credits per minute instead of 1) and review both versions side by side. Timestamps do not change with translation, but translated lines can be longer, so give them a read.
What does it cost?
1 credit per started minute for transcription, 2 credits per minute when you add translation. A 12-minute episode is 12 credits as a transcript and 24 with a translation. New accounts start with 5 free credits.
Why is there a 20 minute limit?
Long recordings take longer to review and are heavier to process. Split longer files and process the parts.
Can I fix the transcript before exporting?
Yes. Edit any line, and adjust its start and end times. Both are free.