Media and production
Interview and documentary rushes transcribed and timecoded, so editors cut from a paper edit instead of scrubbing through hours of footage. Part of our wider media language services.
Transcription Services
Timecoded transcripts and SRT/VTT files from your footage, in 300+ languages, ready for editing or upload.
Video transcription converts the speech in a video into a written document, timecoded so every line can be traced back to a point in the footage. That document is not a subtitle file: it is text you read, search and quote from, while subtitles are cues timed to appear on screen. Prism Linguistics transcribes video in more than 300 languages.
A transcript makes a video far more useful. It gives editors something to cut from, gives marketers quotes and show notes, makes the content searchable, and is the sensible first step toward subtitles that open the video up to a wider audience.
These two get ordered interchangeably and they are not the same deliverable. Getting it right at the quote stage saves everyone a wasted week.
If you know the footage eventually needs subtitles in four languages, it is tempting to skip straight there. We would usually still start with the transcript.
It is the cheapest point at which to catch problems: a misheard drug name is a five-second fix in a Word document and a much bigger nuisance once it has been cued, timed and translated four ways. It is also what everyone else on the project needs, from the producer approving content to the legal reviewer checking what was said. And one approved transcript feeds every downstream language, so the translation teams work from a signed-off source rather than four people separately interpreting the audio.
There is a fourth reason that only shows up later. Edits move. If the cut is still changing, timing subtitles is wasted effort, whereas a transcript of the content survives a recut almost intact.
Timecodes are what make a transcript usable against footage rather than merely readable. Ours default to hours, minutes and seconds counted from the start of the file, placed at every change of speaker, because that is what most people want and nobody has to explain it.
Other conventions are no problem if you say so with the brief. Where the footage carries embedded timecode from the camera or the edit, we match it rather than counting from zero, including drop-frame where it applies. Editors working to a paper edit usually want them more frequently, often every thirty seconds. Legal work wants them wherever something material is said. Speaker labels work the same way. Send a cast list with the footage and the transcript carries real names and roles rather than anonymous numbering, which is the difference between a file you can search in six months and one you have to re-watch to understand.
Send MP4, MOV, AVI, MKV or most things a camera or an edit suite produces. For large files we will set up a secure upload link, and we can work equally well from an unlisted YouTube or Vimeo link, a Dropbox or SharePoint folder, or a shared drive. Broadcast masters are not necessary and often slow things down: clean audio matters far more than picture resolution, and a compressed proxy with good sound beats a pristine file with the microphone six feet from the speaker.
What comes back is a Word document by default, or PDF, plain text, or your organisation's own template. Where the job is heading toward captions, we can supply subtitle-ready SRT, VTT or SBV files alongside the transcript so the editor has both in hand.
The deliverable barely changes; what people do with it changes completely.
Interview and documentary rushes transcribed and timecoded, so editors cut from a paper edit instead of scrubbing through hours of footage. Part of our wider media language services.
Course video transcribed for downloadable transcripts, accessible learning materials and the caption files that usually follow.
Webinars, town halls and brand films transcribed for the record, for repurposing into written content, and for accessibility.
YouTube and social footage, with transcripts that give search engines something to index and caption files that hold attention in a silent feed.
Recorded video interviews and observational data transcribed and timecoded for coding and analysis, alongside our audio transcription work.
Video in 300+ languages, transcribed in the original, translated into English, or both with the timing intact.
Foreign-language footage takes two distinct skills, worth separating when you brief it. Someone who speaks the language on screen has to listen and write down what was said. Only then can it be translated.
Running both stages with us keeps the timecodes aligned, which sounds like a small thing until you are trying to match an English quotation back to a moment in the rushes. You can have the original-language transcript, an English translation, or both in parallel columns. Documentary teams and legal teams almost always want both, for opposite reasons: one wants to check the nuance, the other wants to prove the wording.
The translation side is handled by the same people behind our translation services, so a native speaker of the target language does the work and a second linguist reviews it. Community-language footage needing Arabic translation services for a local authority, or interview material needing translation from Polish into English for a production company, both run this way. If the footage is going to end up with text on screen rather than in a document, subtitling and closed captions picks up where the transcript stops.
The transcript often earns its keep long after the project it was ordered for.
Three steps, and the more you tell us at step one the tighter the quote.
Upload through the quote form, share a private link, or ask us for a secure upload. Tell us the total runtime, the language spoken, roughly how many people appear, and the deadline you are working to.
We agree how often you want timecodes, whether they run from zero or match embedded timecode, what speaker names to use, and whether you want subtitle-ready files alongside the transcript. Then a fixed price and a date.
A transcriber works through the footage, and the transcript is checked against the video before it goes out, with anything genuinely inaudible flagged with a timecode rather than guessed. Translation runs alongside it if you asked for it.
Quoted per job, not off a rate card, because two hours of footage can mean two very different amounts of work.
Deadline sits across all of it. Genuinely urgent work carries a premium, and if a day's wait would save you money we will say so. Our pricing page sets out how we price language work generally; for footage, a short sample tells us more than any rate card. Send it with the runtime and the quote comes back fixed.
If there is no picture involved, our audio transcription page covers interviews, focus groups, dictation and meeting recordings, including the verbatim and intelligent verbatim styles. If the end product is text appearing on screen for viewers, that is subtitling and closed captions, which adds cueing, condensing and reading-speed work on top of the transcript. Everything else we do is on the services page.
Send us a link or the file, and tell us whether you want a transcript, subtitle files or both. We will quote and confirm a turnaround.