AI subtitles and captions, with a native linguist on every job
Send your videos, and get subtitles in other languages, same-language captions, and transcripts from one workflow. AI transcribes, times, and translates the speech. Native linguists check the transcript against the audio and work on the subtitles, at the depth your service level sets. Before any file ships, automatic checks flag every subtitle that is too fast to read, too long for its line, or on screen too briefly or for too long.
Product tour video · English → ArabicExample
Subtitle 42From 1 minute 15.32 seconds to 1 minute 17.12 seconds1.80 s
- Speech
- Need an older version? Open History.
- AI draft
-
هل تحتاج إلى نسخة أقدم؟ افتح السجل.
Flagged: reading speed 19.4 characters per second (your spec: up to 17)
- Edit
- Linguist
النسخ السابقة في السجل.
In English: Previous versions are in History.
- Checks
-
- Reading speed: 12.8 characters per second (up to 17)
- Line length: 23 characters (up to 42)
- Lines: 1 (up to 2)
- Duration: 1.80 s (0.83 to 7 s)
- Issue log
- Subtitle 42 · Client requirements · Minor · Reading speed over your spec. The linguist condensed the line without losing the meaning and kept السجل for History, the label in your Arabic interface.
What you get
- Subtitles in other languages. Translated subtitles, timed to the speech and condensed where a speaker is too fast to read in full, so viewers can follow without pausing.
- Captions in the same language. The speech on screen in the language spoken, for viewers who watch with the sound off or prefer to read along.
- On-screen text. Titles, labels, and other text in the picture that viewers need, translated as subtitles when your spec asks for it.
- Captions for viewers who are deaf or hard of hearing (SDH). Same-language captions that also name the speaker when it isn’t clear who is talking, describe the sounds and music that matter, such as [door slams] or [upbeat music], and mark song lyrics.
- Transcripts. The full text of the speech, with timecodes, as a Word or plain-text file you can search, quote, or send for translation.
- An issue log. What the automatic checks and the linguist found and fixed in each file.
All subtitles, captions, and transcripts come from the same checked transcript, so names, numbers, and terms match across the languages you order.
What this service doesn’t cover
We work on recorded video and audio. We don’t caption live events.
Need Arabic voice-over or dubbing? Our multimedia team works at locstars.com (external site).
Where and how you publish a video decides whether it needs captions and which rules the captions must meet. Your team confirms that, and our files follow the spec you send us.
How it works
Every step has a named owner, and a native linguist works on every job.
- Intake and spec. You send the video or audio, or links to it, and tell us the languages and what you need: subtitles, captions, captions for viewers who are deaf or hard of hearing, or transcripts. If you have a script, a transcript, or subtitle files, send them too. Before work starts, we agree on the spec: reading speed, and whether spaces count toward it; characters per line; lines per subtitle; minimum and maximum duration; gaps between subtitles; the frame rate; how closely captions follow the speech, word for word or without fillers and false starts; how speakers, sounds, and on-screen text are shown; and the file format. We use your style guide or your platform’s subtitle spec. If you have neither, we propose a spec for you to approve.
- Names and terms. From your glossary, script, and published content, agents collect the names, product terms, and places that appear in your video, each with its source, and a native linguist approves the list.
- Transcription and timing. AI transcribes the speech and times each line to the audio. If you send a script or a transcript, we align that text to the audio instead.
- Transcript check. A native linguist of the spoken language checks the transcript against the audio, at the depth your service level sets: names, numbers, terms, and who is speaking. Speech that can’t be made out is marked, not guessed.
- Translation. AI translates the checked transcript sentence by sentence, with the context of the whole transcript, against your glossary and style rules. It then fits each sentence to the timed subtitles, merging or splitting subtitles where the target language needs it, within the reading-speed and line-length limits of your spec.
- Linguist work. A native linguist of each target language works on the subtitles your service level covers, with the video playing: meaning and tone, condensing speech that is too fast to read in full, line breaks, and timing, including around shot changes. For captions for viewers who are deaf or hard of hearing, the linguist also adds the speaker labels and sound descriptions.
- Quality gate and delivery. Automatic checks run on every subtitle in every file: reading speed in characters per second, characters per line, lines per subtitle, minimum and maximum duration, gaps and overlaps between subtitles, and your approved terms. The linguist settles every flag by fixing it or recording why it stays, and if a file fails the gate, it does not ship.
This is our AI cycle applied to video: AI does the volume work of transcribing, timing, and translating, and a native linguist decides every finding.
How deep the linguist’s work goes
Subtitles and captions run on our three service levels. Your quote names the level for each type of video.
| Service level | Transcript | Subtitles and captions |
|---|---|---|
| AI + Check | Every line our agents or checks flag, plus a random sample, checked against the audio | Every subtitle our agents or checks flag, plus a random sample, checked and fixed by a native linguist |
| AI + Edit | Every line, checked against the audio | Every subtitle, edited by a native linguist, then a sample checked by a second linguist |
| AI + Edit + Review | Every line, checked against the audio and then reviewed by a second linguist | Every subtitle, edited and then reviewed against the video by a second linguist |
AI + Check fits internal and archive video, such as recorded meetings and internal training. At AI + Check, subtitles that are not flagged or sampled keep the AI’s timing, which follows the speech but not shot changes. AI + Edit is the usual level for product, customer training, and support video. AI + Edit + Review fits campaigns and brand video.
Captions for viewers who are deaf or hard of hearing run at AI + Edit or above, so a native linguist checks every caption, speaker label, and sound description. Video with regulated or high-risk content, such as health, legal, or financial information, runs at AI + Edit + Review or without AI, never at AI + Check.
Arabic subtitles and captions
- Right to left on screen. Video players handle right-to-left text differently. In some, the period or question mark at the end of an Arabic subtitle appears at the wrong end of the line, and numbers or English names inside a line appear out of order. Tell us where your video will play, and we follow that platform’s spec for text direction where it has one, and add direction marks where your player needs them. The linguist checks the subtitles on screen, against the video, not only as text. Before a large delivery, we suggest you test one file in your own player.
- Numbers. Western digits, such as 25, or Arabic-Indic digits, such as ٢٥: your spec sets which, and whether numbers from one to ten are written as words, and we use the same choices in every file.
- Line breaks. Linguists break lines where Arabic grammar allows, not where the character count runs out: a noun stays with its adjective and with the noun that completes a possessive phrase, and a preposition stays with its object.
- Captions for viewers who are deaf or hard of hearing. Speaker labels and sound descriptions are in Modern Standard Arabic by default, in brackets, such as [موسيقى هادئة], and song lyrics are marked with ♪.
- Dialect. Subtitles are in Modern Standard Arabic (MSA) by default. It is the written standard that viewers read across markets. For marketing, social, and other user-facing video, Egyptian, Gulf, Levantine, and Maghrebi Arabic are available on request. When the speech itself is in a dialect, you choose whether the captions follow the dialect as spoken or render it in MSA.
Languages and dialects
We subtitle in the languages we list, including Modern Standard Arabic and, on request, Egyptian, Gulf, Levantine, and Maghrebi Arabic. Other languages on request: we confirm we can run them before we quote.
Subtitles run between English and each listed language, and between Arabic and each of the others, in both directions. Captions and transcripts are in the language spoken in your video, when it is one of the languages we list.
What you send and what you get back
You send
- Video or audio files in MP4, MOV, MP3, WAV, or M4A. Our form takes up to 100 MB in total, which is often only a minute or two of full-HD video, so for most videos, send a link or a smaller copy: a 720p MP4 is enough for timing. Send the final cut, at the frame rate you publish, so the timing matches your video.
- Subtitle files you already have, in SubRip (SRT) or WebVTT (VTT), if you want them translated
- A script or a transcript, if you have one
- Your glossary, style guide, or platform subtitle spec, if you have them
- The target languages, and where the video will play
You get back
- Subtitle and caption files in SRT or WebVTT, in UTF-8, one for each language. If your platform needs another format, tell us, and we confirm before we quote.
- Transcripts with timecodes, as Word (DOCX) or plain-text files
- The quality-gate result for each file
- An issue log, plus the quality reports your service level includes
- Glossary changes and a query log at AI + Edit and AI + Edit + Review
- Your translation memory (TM), updated, with a Translation Memory eXchange (TMX) file on request
How we quote subtitles and captions
Video has no word count until it is transcribed, so we quote per minute of video or audio. The quote depends on what each minute needs: transcription, the language or dialect spoken, the number of target languages, the type of captions, how fast and dense the speech is, the audio quality, and the service level. A script, a transcript, or subtitle files you already have can reduce the work, and the quote shows where they do. The quote states the basis, and we confirm any change with you before work continues.
Your video and your data
- We sign your NDA, and your data processing agreement (DPA) if you have one, before you share files. If a clause doesn’t match how we work, we agree on the change with you before work starts.
- Video and audio often include people’s faces, voices, and names. Send only video you have the right to share with us, and tell us before you send it if it includes customers, patients, staff, or children, so we can agree on how it is handled. We don’t use face or voice recognition to identify people in your video.
- When we use AI on your content, we use only business AI services whose terms bar training on it. We don’t use your content to train or tune our own AI tools.
- Prefer no AI? Native linguists then transcribe, time, and translate your video, with no AI or agents at any step on our side, and the quote says so.
- We delete your files on request, unless the law requires us to keep them. Files you upload through our quote form are deleted automatically from our form storage 90 days after upload.
Start with 3 minutes, free
Send a video file under 100 MB, or a download link to one, such as a cloud-storage link, and tell us which 3 minutes to use. We subtitle up to 3 minutes into one target language, or caption them in the language spoken, at AI + Edit: a native linguist edits every subtitle. If you choose No AI on the form, or the video has regulated or high-risk content, such as health, legal, or financial information, a native linguist subtitles the 3 minutes with no AI on our side. You get the subtitle file in SRT or WebVTT, the quality-gate result, and the issue log. The sample covers only the minutes we worked on. Send only video you have the right to share, and if you can, choose 3 minutes with no personal data on screen. For the free sample, please don’t send video of patients, customers, or children, or video with health or other sensitive information about a person. One free subtitle sample per company.
On the quote form, choose “Video subtitles or captions” and the free subtitle sample, and add your video as a file or a link. Under Anything else, tell us which 3 minutes to use. We confirm the details by email before we start.
Questions about subtitles and captions
What is the difference between subtitles and captions?
Subtitles put the speech, and on-screen text where your spec asks for it, into another language, for viewers who can hear the audio but don’t understand the language. Captions put the speech on screen in the language spoken. Captions for viewers who are deaf or hard of hearing, often called SDH, also name speakers and describe the sounds that matter. We make all three from the same checked transcript.
Can you translate subtitles we already have?
Yes. Send your SRT or WebVTT files with the video. A native linguist of the spoken language checks them against the audio first, at the depth your service level sets, and we keep your timing wherever it works for the target language within your spec.
Do you subtitle Arabic dialects?
Yes. Subtitles are in Modern Standard Arabic by default, and for marketing, social, and other user-facing video, Egyptian, Gulf, Levantine, and Maghrebi Arabic are available on request. Tell us which countries your viewers are in, and we confirm the variant before we quote. When the speech is in a dialect, you choose whether captions follow the dialect or render it in Modern Standard Arabic.
How do we send large video files?
Our quote form takes up to 10 files and 100 MB in total, in MP4, MOV, MP3, WAV, or M4A. That is often only a minute or two of full-HD video, so for most videos, share a link, such as a cloud-storage link, or send a smaller copy. Please send the final cut, at the frame rate you publish. Please don’t put passwords in our forms: if a link needs access, we arrange it by email.
Which file formats do you deliver?
SRT and WebVTT for subtitles and captions, and Word or plain text for transcripts. If your platform needs another format, tell us, and we confirm before we quote.
Can we choose no AI?
Yes. Native linguists then transcribe, time, and translate your video, with no AI or agents at any step on our side, and your quote says so. Video with regulated or high-risk content runs at AI + Edit + Review or without AI.