How to transcribe a video to text
A step-by-step guide to transcribing a video to text: pick the right file, get clean audio, run the transcript, and fix names before exporting SRT or VTT.
Free sign-up, 60 min/month · 99+ languages · Up to 500 MB per file
What is Video Transcription Workflow?
Transcribing a video to text means turning the speech in a recording into a written document while keeping each line tied to its timecode. The video container does not need to be converted first: MP4, MOV, and WebM files can be read directly, the spoken language is detected, and the result is an editable transcript. The same pass can also produce SRT or VTT captions for the clip.
Why choose Video Transcription Workflow?
This guide covers the whole path rather than one button: which files work, when audio quality matters, how to handle several speakers, and what to check before exporting. It is written for people who have a video and need searchable text, not for editors working on a timeline.
Retyping a Video One Sentence at a Time?
A five-minute clip can take an hour to transcribe by hand because every pause means another rewind. Watching the timeline to catch a name or a number is slower than reading it. A transcript that keeps timecodes lets the file be searched like a document, a line be quoted without guessing where it came from, and the same pass be turned into captions. The steps below keep the video intact and work from a copy the uploader owns.
What makes Video Transcription Workflow different?
The guide treats video as the source file: no audio extraction, no format conversion, and no second copy of the recording to manage. One upload, one pass.
Works on MP4, MOV, and WebM files
Keeps a timecode on every line
Exports text, SRT, or VTT
Formats, languages, limits
- Languages
- 99
- Detected automatically. Set it by hand when a recording mixes languages or carries a heavy accent.
- File types
- 13
- MP3, M4A, WAV, AAC, MP4, MOV, WebM and more. Video is processed directly — no separate audio step.
- Per file
- 500 MB
- Large files upload straight to storage with a visible progress bar, up to 500 MB.
- Exports
- TXT · SRT · VTT
- Plain text, SRT and VTT subtitles, ready for a document, an editor, or a caption track.
How to use Video Transcription Workflow in 3 steps
Check the file and the audio
Use a video you own or have permission to use; a clear voice track matters more than resolution.
Upload and set the language
Add the video, let language detection run, and turn on speaker separation when more than one person speaks.
Correct names, then export
Read through for names and numbers, then export plain text, SRT, or VTT.
Video Transcription Workflow
Try the steps on a video you already have. Sign up free and get 60 minutes a month, no card required.
Free sign-up, 60 minutes a month, no credit card. Export when you are done.
Transcribe a videoFrequently asked questions about Video Transcription Workflow
Which video files can be transcribed?
MP4, MOV, WebM, M4A, WAV, and MP3 are accepted directly, so common phone, camera, and screen recordings work without conversion. A file whose audio codec the browser cannot preview may still upload; if it fails, export the sound track as MP3 or M4A and upload that instead.
Do I need to extract the audio first?
No. The video is read as it is, and the transcript still carries the timecodes of the original clip. Extracting audio is a fallback for an unusual codec, not a required step. Keeping one file also avoids a second copy of a recording that may be sensitive.
How long does transcription take?
A short clip usually finishes in less time than it takes to listen to it, while an hour-long recording takes proportionally longer. Processing runs in the background, so the tab can be closed and the transcript opened later. Two jobs can run at once, and a third waits in the queue.
Can I transcribe a video with several speakers?
Yes. Speaker separation can be switched on before the job starts, and each change of voice begins a labelled turn. Labels can be renamed to real names in the editor. Recordings with one microphone per person separate best; overlapping speech may be merged into a single turn.
What should I check in the transcript?
Names, numbers, product terms, and the boundaries between speakers are worth a pass. Clean audio needs little correction, while noisy or accented recordings benefit from reading against the timecodes. Corrections are saved with the transcript, so the export holds the corrected text.
Can the transcript become subtitles?
Yes. The same timecoded transcript can be exported as SRT or VTT and imported into a video editor or player. Line breaks may need a light edit to meet caption length rules, but the timing is already in place, so the caption file does not start from an empty timeline.
Is the video kept after transcription?
The file is uploaded to storage only so the recognizer can read it, and it is deleted when the job finishes or fails. The transcript stays in the account until it is deleted. Audio is not used to train models, and the recognition provider processes it only on Dictawave's instructions.
Can a video from YouTube or TikTok be transcribed?
Not by downloading it. Dictawave does not download or extract content from third-party platforms. If the uploader owns the file or has permission to use it, the video can be uploaded directly and transcribed like any other recording.