Professional Subtitling Services in 150+ Languages

Subtitling services turn the dialogue in a video into timed lines of text, then hand you a subtitle file your player can read. Espresso Translations subtitles video in 150+ languages from $5 per video minute, delivers standard projects in 2 to 3 working days with 24-hour express available, and has a second linguist revise the subtitle translation under ISO 17100:2015 before the file comes back. Buyers who have not ordered this before usually begin by asking for the wrong product: subtitle and caption are different things built for different viewers, and ordering a caption track when the audience does not speak the language is the commonest reason a finished delivery has to be done twice. So this page settles that first, then gives you the price, the formats, and the five things worth checking before you hire a subtitling company. Subtitling and translation are one job here, not two suppliers.

ISO 17100:2015 Certified

Native Subtitlers

From $5 Per Video Minute

Second-Linguist Revision

Subtitles in 150+ Languages

Free Quote in 1 Hour

Espresso Translations reviews on Google
Espresso Translations reviews on Trustpilot
Espresso Translations ISO 17100:2015 certification, the translation-services standard

Send a Video for a Free Subtitling Quote

A Project Manager replies within 1 hour

Forbes, a client of Espresso Translations
Transcription and subtitling services for IMDb from Espresso Translations
Jaguar, a client of Espresso Translations
Universal, a client of Espresso Translations
National Geographic, a client of Espresso Translations
EY, a client of Espresso Translations
Sixt, a client of Espresso Translations
Amazon, a client of Espresso Translations
EDF, a client of Espresso Translations
FREENOW, a client of Espresso Translations

What are subtitling services? Subtitling services translate and time the dialogue in a video, then deliver it as a subtitle file — SRT, VTT, STL or TTML — that the player loads over the picture. What separates a subtitling service from an automatic caption button is that somebody decides what each line has to lose: a subtitle has to sit inside a line length and a reading speed as well as carry the meaning, so the text is condensed rather than transcribed. Espresso Translations works in 150+ languages with native speakers of the target language, from $5 per video minute, and a second linguist revises the subtitle translation under ISO 17100:2015. Standard delivery is 2 to 3 working days, 24-hour express is available, and an itemized quote is free within 1 hour. One thing to settle before you order: if your audience can hear the audio and simply does not speak the language, you want subtitles; if they cannot hear it, you want captions or SDH, and the two files are not interchangeable.

What Do Subtitling Services Include, and Which Languages and Content Do They Support?


A professional subtitling service delivers three things: the translated text, the timing that holds it against the picture, and a subtitle file in the format your platform reads. Espresso covers all three from $5 per video minute. Buyers ask for subtitle services, subtitling services and video subtitling and mean one purchase by all three: a translated, timed file built for a named platform. Send only a video and the timed transcript is made as part of the job; send an SRT or a VTT you already hold and the existing cue timings are kept while only the text is translated. Accepted source material includes MP4, MOV, AVI and WMV alongside SRT, VTT, TTML and ASS, and delivery can be a separate subtitle file, a burned-in version, or both from one set of timings.

The work goes to professional subtitlers who are native speakers of the target language and who work in audiovisual content, not to document translators and not to a bilingual colleague on your own staff. Reading two languages is a different skill from writing a line that fits the two seconds it holds the screen. A subtitler makes an editorial decision in every cue — which clause survives, which one can be implied, whether a joke needs the cultural adaptation of being replaced rather than rendered — and that decision is invisible when it is made well and obvious to every viewer who speaks the language when it is not.

A subtitle is not a transcript, and that is the sharpest difference between what a subtitling service produces and what a speech-recognition tool produces. Speech runs faster than reading, so a verbatim line either outruns the viewer or stays on screen after the shot has cut. The subtitler compresses instead: the sense, the register and the tone stay, the redundant words go. That compression is also why translated subtitles built from a verbatim caption track usually read badly — the source lines were never cut for the screen in the first place.

Every subtitling project runs under a signed NDA with encrypted transfer and access limited to the people working on the file, which is what unreleased film, pre-launch product video and internal compliance material need. One Project Manager runs the job from upload to delivery, confirms the output format against the platform specification before work starts, and answers the technical questions. The subtitle translation itself is revised by a second linguist under ISO 17100:2015, the translation-services standard Espresso Translations holds certification to.

Subtitles, Captions or SDH: Which One Does Your Video Need?


Order the wrong one of these and the file is useless to the audience it was bought for. The three products look alike in a player and are built for different viewers: subtitles for someone who can hear the audio and does not speak the language, closed captions for someone who cannot hear it, SDH for a deaf or hard-of-hearing viewer on a platform that carries no separate caption track. US accessibility rules ask for captions, not for translation — Section 508 requires public-facing electronic content to meet WCAG 2.0 Level A and AA, and the captions criterion for prerecorded media sits at Level A — so a compliance requirement and a foreign-language audience are two separate reasons to buy, needing two different files.

Subtitles, closed captions, SDH, burned-in and forced narrative compared
TypeWhat the text carriesWho it is for
SubtitlesDialogue translated into another language, condensed to fit the screenA viewer who can hear the audio track but does not speak the original language
Closed captionsDialogue in the original language, close to verbatim, plus the non-speech sound that carries meaningA viewer who cannot hear the audio, or who is watching with the sound off
SDHDialogue plus speaker labels, sound effects and music cues, in the original language or translatedA deaf or hard-of-hearing viewer on a platform that carries no separate caption track
Open subtitles, also called burned-in subtitlesText rendered permanently into the picture, with no track the viewer can switch offSocial platforms that strip a subtitle file out of the upload, and screens with no caption support
Forced narrative subtitlesOnly the lines a viewer must read: dialogue spoken in another language inside the film, and signage or writing left untranslated on screenA viewer watching in the main audio language who would otherwise miss those lines
Uploading a video and naming the target languages for a free subtitling quote
A second linguist revising a translated subtitle file against the source video
Project Manager confirming subtitle format, reading speed and delivery date
Finished subtitle files delivered in SRT, VTT, STL and TTML

What Is a Subtitle?

A subtitle is a line of text, timed to the dialogue, telling a viewer what is being said. It sits at the foot of the picture, appears when the line starts and clears when it ends, and it assumes the viewer can hear the audio track. Subtitles are usually in a language other than the one being spoken, which is the point of them: a non-native audience follows the film without the audio being replaced. They can also be in the source language, for a viewer who wants the words as well as the sound, or for one watching in a noisy place. What a subtitle does not do is report every sound in the scene — that is a caption’s job, and the two are routinely confused by the people buying them.

What Is Captioning, and Why Is It Not the Same Thing?

Captioning writes down the dialogue in the language it was spoken in, close to verbatim, for a viewer who cannot hear the audio. It also carries the sound that matters — a door, a phone, the music cue that tells you the scene has turned — because a viewer who cannot hear would otherwise lose it. A caption does not translate, so it does nothing at all for an audience that cannot understand the language. That distinction decides what you order, and it is also the distinction US law draws: 47 CFR 79.4 requires closed captions on full-length internet video only where the programming was published or exhibited on television in the United States with captions, and it expressly excludes consumer-generated media and clips. Nothing in that rule, and nothing in WCAG 2.0’s captions criteria, asks for a second language. Captioning is often an accessibility obligation; translation is a commercial decision about reach. Most projects that buy both buy them for those two unrelated reasons, which is also why subtitling and closed captioning are quoted as separate work even when one video needs both.

How Do Subtitles Actually Work: the File, the Cues and the Timing?

A subtitle file is a numbered list of cues, and every cue holds a start time, an end time and one or two lines of text. In an SRT the cues are numbered from 1, the times run hours:minutes:seconds with a comma before the milliseconds and an arrow between them — 00:02:16,612 –> 00:02:19,376 — and a blank line closes each cue. A WebVTT file opens with the word WEBVTT and uses a period before the milliseconds instead of a comma. Nothing is written into the video itself: the player reads the cue list and paints each line over the picture at the moment it is told to, which is why the viewer can switch subtitles off or change language. Setting those in and out points against the picture is the step subtitlers call spotting, and the timecoding is done to the shot rather than to the waveform, because a line straddling a cut reads as an error even when every word in it is right. Published specifications put numbers on the rest. Netflix’s English timed-text guide allows 42 characters per line and a maximum of two lines, a reading speed up to 20 characters per second for adult programs and 17 for children’s, and a cue duration from five-sixths of a second up to seven seconds. Those are one distributor’s rules rather than a universal standard, but they are the right order of magnitude for any online video, and they are why a translation that reads perfectly in a document still has to be cut before it can be a subtitle.

When Do You Need SDH, or Subtitles Burned Into the Picture?

SDH — subtitles for the deaf and hard of hearing — is the file to ask for when the platform carries one subtitle track and your audience includes viewers who cannot hear. It does a caption’s job inside a subtitle track: dialogue plus speaker labels plus the sounds that carry meaning, in the source language or translated for an international audience. Burned-in subtitles are also called open subtitles, and burned-in captions are called open captions, because neither kind can be closed. Both answer a different problem — a social platform that discards a separate subtitle file on upload, or a screen with no caption support at all. The trade is permanence: the viewer cannot turn them off, cannot change language, and a new language means a new video export rather than a new file. Audio description, which narrates what is happening on screen for blind and partially sighted viewers, is a separate accessibility track again and is not part of any subtitle file.

How Do You Judge a Subtitling Company Before You Hire One?


Five things separate a subtitling company that can do the job from one that hands back a file you have to fix, and you can check all five before you spend anything. They are the transcription behind the subtitles, what is actually being checked and by whom, readability, how well a line survives being cut, and whether you can get the same timings in every format you will need. Ask about them in that order. A subtitling service provider who answers all five concretely is describing a real workflow; one who answers in adjectives is describing a sales page.

Transcription Background: Who Makes the Timed Transcript?

Subtitling starts from a transcript of the audio, timed against the picture, and if you do not have one it has to be made before anything can be translated. Converting audio or video into timed text is the step everything after it rests on. This is where a thin supplier shows first. A clean single-speaker recording is easy; two people talking over each other, a site visit with machinery running, a conference microphone picking up the hall — those need somebody who transcribes for a living, and an error there propagates into every language you order afterwards. So ask who does that step, and whether a human listens to the audio materials or only edits what a tool produced. Send a representative five minutes rather than the easy opening, because a quote and a sample are only as honest as the material they were made from.

One thing to be precise about, because the market is not: ISO 17100:2015 is a translation-services standard whose scope does not reach transcription at all. It also leaves out interpreting and raw machine-translation output. The certificate therefore says something about the subtitle translation and nothing whatever about the transcript — we do not describe a transcript as certified, and a supplier who does is claiming something the standard cannot support. What it does say is worth having. Under ISO 17100:2015 a second person has to revise the translation, which is why every subtitle file leaving here has been through a translator and then a reviser, rather than through one person twice.

Accuracy: What Is Being Checked, and by Whom?

Nobody can hand you a meaningful accuracy percentage for a subtitle file, and a supplier offering one is quoting a number that was never measured. There is no standard test, no graded sample and no agreed unit: a mistimed line break, a literal idiom and a misheard proper noun are all errors, and none of them is a fraction of a word. What you can check is the process. Ask who revises the translation, whether the reviser works against the source audio and the picture or only against the translated text, and what happens to the file after the language work is finished. A bilingual revision against the video catches the errors that matter in subtitling, because most of them are errors of context rather than of vocabulary — who is being addressed, which of two possible meanings the shot settles, whether a pronoun belongs to the speaker or to the person on screen. Accurate translations of subtitles are won in revision, not claimed in a quote, and that is the part of professional translation a buyer can actually inspect.

At Espresso the sequence is fixed. A subtitler who is a native speaker of the target language translates and times the file. A second linguist then revises it against the source video under ISO 17100:2015. A technical pass checks the file itself — cue numbering, reading speed, line length, character encoding, and whether the format matches the platform you named. Linguistic accuracy and technical validity are checked separately on purpose, because a file can be right on the first and broken on the second, and quality control that only reads the words will pass the second kind of fault straight through to the upload that fails.

Readability: Characters Per Line, Lines on Screen, Reading Speed

Readability is measurable, which makes it the easiest of the five to check. Three numbers govern it: how many characters fit on a line, how many lines appear at once, and how fast the text moves. Netflix publishes 42 characters per line, two lines maximum, and a reading speed up to 20 characters per second for adult programs and 17 for children’s; most broadcast and streaming specifications sit in that region. Ask a supplier which numbers they build to, and what happens to a line that exceeds them. The answer tells you whether anybody is measuring at all, because this failure is invisible in a text document and obvious the moment the file is played.

The pressure on those numbers comes from the languages. A line that fits comfortably in English frequently does not fit once it is translated, and the subtitler has to recover the space without losing the sense, which is editorial work and not a settings change. Position matters too: a subtitle has to clear on-screen text, burned-in titles and anything else at the foot of the frame, and in broadcast formats the position is carried inside the file rather than left to the player. None of this shows up in a word-count quote, which is one reason subtitling is priced per video minute.

Simplicity: Does the Line Survive Being Cut to Fit?

Simplicity is what is left when a subtitle has been cut to its time. One idea per cue, the clause order the eye expects, no sub-clause that sends the viewer back to the start of the line — the reader is watching a picture at the same time, and anything needing a second pass costs them the shot. Typography follows the same rule: a plain font, enough contrast against the picture, and none of the colored or decorative styling a styled format makes possible and almost nothing needs.

You can test this before you buy. Ask for a sample minute of your own video rather than something from the supplier’s reel, and read it with the sound off. Follow the scene without rewinding and the condensation is working. Find yourself reading rather than watching and the lines are too long, or they are being held too briefly — and no amount of linguistic quality will rescue a file that reads like a transcript dropped into cues.

Flexible Formats: Can You Get the Same Timings in Every Format You Need?

One video usually ends up on more than one platform, so ask up front which formats a subtitling provider can deliver from a single set of timings. SRT holds plain text and time codes and is what most web players and social platforms want; VTT adds basic styling and is the HTML5 format; TTML and DFXP carry timing, positioning and styling for streaming delivery; EBU-STL and SCC are the formats broadcast content is delivered in, and broadcast compliance is where positioning stops being optional. YouTube alone accepts SRT and SubViewer as its basic formats, WebVTT, TTML and DFXP as advanced ones, and SCC and EBU-STL among the broadcast formats it will ingest; Vimeo and the usual web players take SRT or VTT. Converting between formats is cheap once the spotting is right and expensive when it is not, which is why the question is about the timings rather than about the file extension.

Flexible delivery also means burned-in output where a platform demands it, SDH alongside the translated track where the audience needs both, and one terminology held across multiple languages in the same set. If you know a video is going to a social feed, a learning platform and a broadcaster, say all three at the quote stage. A supplier who builds the timings once for all three costs less than one who rebuilds them per platform, and you will not find out which you hired until the second delivery.

What Do Subtitling Services Cost, and What Comes Back?


Subtitling starts at $5 per video minute, and the unit is the running time of the video rather than a word count — a 12-minute film is quoted on 12 minutes. That floor covers the timed transcript where you do not have one, the spotting, the translation, the second-linguist revision, and the finished subtitle file in the format you asked for. Nothing on that list is metered separately: no charge per cue, no charge for a second output format, no subscription and no monthly minimum.

Four things move a quote above the floor. How much dialogue the video content carries, because a ten-minute interview and a ten-minute montage are not the same amount of work even though they bill on the same ten minutes. The language pair, because a common combination has more available subtitlers than a rare one: language coverage is not the same thing as depth, and less common languages get scheduled rather than simply quoted when a project covers several language pairs at once. The output format, because a plain SRT is less work than a styled broadcast package carrying positioning. And the deadline, because a fast turnaround has to be resourced on the day it is asked for. Terminology sits across all four: legal, medical and engineering content goes to a subtitler who works in that field, and that is a different person from the one who handles a product launch.

If you already hold a transcript or a subtitle file, the job changes shape. Translating an existing SRT or VTT is quoted on its word count at $0.10 per word, the time codes are left alone, and only the lines that no longer fit the reading speed are re-timed. That route suits dialogue-heavy corporate videos, training videos and e-learning libraries, where a script already exists and the transcription step is the part you do not need to pay for twice.

There is no hourly rate for any of this, and the per-minute price is not a disguised one — it is a unit price for a deliverable. The work ratio behind it is worth knowing when you plan a schedule, because a minute of dense two-speaker dialogue takes several times as long to spot and condense as a minute of narration over a montage, which is why a dialogue-heavy video is quoted higher per minute instead of simply arriving later. Where a recording has to become a document rather than a subtitle track, for a court file or an immigration application, that is certified translation at $30 per page up to 250 words, or sworn translation at $55 per page delivered in 7 to 14 working days where a sworn translator’s signature is required.

Turnaround time is agreed before anything starts. Standard subtitling projects are delivered in 2 to 3 working days, with 24-hour express available when a release date is fixed; a feature-length film, a series, or one video going into eight languages at once takes longer, and the date is written into the quote rather than estimated afterwards. Send the video — or a representative five minutes of it — with the target languages, the platform and the deadline, and an itemized quote comes back free within 1 hour.

Subtitling prices and delivery times
WorkPriceDelivery
Subtitling a video, including the timed transcript and the spottingFrom $5 per video minute2 to 3 working days
Translating a subtitle file you already holdFrom $0.10 per word2 to 3 working days
Voice-over or a dub of the same videoQuoted per projectAgreed at quote
Itemized quoteFreeWithin 1 hour

Additional Captioning and Subtitling Services

A video rarely needs only one of these. Alongside subtitling we provide transcription services for recordings that stay as text, voiceover where the audience would rather hear the second language than read it, localization services for adapting a whole campaign rather than converting one file, and certified translation services for the occasions when a recording has to become a document. Our captioning services sit inside the same quote as the subtitling, so one video never arrives as two invoices. Interpreting is a live spoken discipline and no part of a subtitling project; we do not offer it.

How Are Subtitles Created, Translated, Synchronized and Quality-Checked?


The process of subtitling runs in four steps, and only the first costs you any time. The Project Manager checks dialogue density, audio quality and the platform specification before quoting, so the price you approve was built on your actual material and not read off a rate card.

78% PLACE ORDER

Step 1: Send the Video, the Languages and the Deadline

Upload the video, plus any transcript or subtitle file you already hold, and say which target languages you need, which platform the file is going to, and when. MP4, MOV, AVI and WMV are all fine, and so is a link where the file is large. The platform matters more than people expect: it decides the format, the reading speed and whether a burned-in version is needed, and naming it now is cheaper than discovering it after delivery.

Step 2: Get an Itemized Quote Within 1 Hour

The Project Manager watches enough of the video to judge dialogue density and audio quality, then builds the quote from the running time, the language pair, the output formats and the deadline. It states the price, the formats and the delivery date, and it arrives free within 1 hour. Nothing starts until you approve it, and if the audio is poor enough to change the price you hear that before the work begins rather than after.

ORIGINAL EN A 文 TRANSLATED AR FR DE ES JA ZH AR CERTIFIED TRANSLATION

Step 3: Spotting, Translation and Second-Linguist Revision

Where no subtitle file exists, a subtitler transcribes the dialogue and spots it, giving every cue an in point and an out point aligned to the shot change wherever the edit allows. Translation then happens inside those cues, condensed to the line length and reading speed your platform specifies. A subtitler has to translate and adapt at once here, so humor, idiom and register are carried across rather than rendered word for word. A second linguist revises the translated file against the source video under ISO 17100:2015. Where you supplied an existing file, the spotting step is skipped and the original timings are kept.

Step 4: Delivery in the Format Your Platform Reads

The translated subtitles come back in the format you asked for — SRT, VTT, STL, TTML, SCC or DFXP — with a burned-in version alongside where a platform strips separate files out of an upload. Every file is validated before it is sent: cue numbering, timing, line length, reading speed, encoding, and the platform’s own specification. Multilingual sets are delivered together, with terminology and formatting held consistent across every language in the batch, so one video does not arrive looking like eight unrelated jobs.

Are Automatic Subtitles Good Enough to Publish?


For some videos, yes, and a page selling subtitling services should say so. A clear single-speaker recording made in a quiet room, for an audience that speaks the language, with no compliance requirement attached, is well served by the automated subtitles a platform generates for nothing. Paying a professional subtitling service for that outcome is money spent on a problem the platforms already solved. The paid work starts where any one of three conditions breaks: the audience does not speak the source language, the audio is harder than one person talking, or the file has to satisfy a published specification. Below that line, manual subtitling buys you nothing the automatic track has not already given you.

What machine output misses is not the words, it is everything around them. Automatic systems translate line by line, so a sentence split across two cues loses its grammar. They time to the waveform rather than to the picture, so lines straddle shot changes. They do not know who is being addressed, cannot tell which of two possible meanings the shot settles, and render idiom and humor literally — visible to every viewer who speaks the target language and to nobody who approved the file. They also have no concept of reading speed, so a verbatim machine caption runs past the 20 characters a second Netflix allows for adult programs without noticing, which is the point where a viewer stops watching and starts reading. Espresso uses human subtitlers throughout, with a second linguist revising the translation, and the subtitling experts on a project treat machine output as something to repair rather than as a draft to trust. Human-made subtitles cost more per minute than a tool and less than a reshoot.

What the Subtitling Tools Market Actually Sells

Worth understanding before you compare a quote against a subscription, because the two are not the same purchase. Subtitling tools are sold as metered minutes or as seats: you supply the audio, the tool returns timed text, and you do the correcting, the condensing and the format work yourself. The platforms bundle a version of this free — YouTube generates automatic captions on upload and lets you replace them by uploading your own subtitle file, which is exactly the deliverable a subtitling service hands over. So the real comparison is not tool against agency, it is your own hours against a quoted price, plus the question of who carries the cost if the file is wrong on a video you have already published. Automation is good at a first timed draft and bad at deciding what to cut, so the content type settles which half of the job can go to a machine. For a weekly internal update the tool wins. For a distributor’s specification, or an audience in a different language nobody on your team reads, it does not.


Subtitling usually starts from a transcript and often ends up beside a voiceover of the same video, and a single language service rarely covers a whole video project, so one of these may be the page you actually wanted. Espresso Translations is a translation agency rather than a subtitling tool, which is why all of it is quoted together.

Transcription Services

Need the words as text rather than as a subtitle track — a research interview, a meeting, a conference panel? Our transcription service handles the recording on its own from $5 per audio minute, video transcription included, and transcription and translation can be ordered as one job.

Video Translation Services: Dubbing and On-Screen Text

Need the whole video localized rather than subtitled — a dub, voice replacement, and the on-screen text as well as the dialogue? See video translation services.

Audio Translation and Voice-Over

Want the second language heard rather than read? Audio translation services covers spoken content, and voice-over is recorded from the translated script and quoted per project.

Audiovisual Translation and Foreign Language Subtitles

Running a project that needs subtitles, dubbing and adaptation together? Audiovisual translation services covers the combined work, and film translation covers feature and documentary content going into international distribution.

E-Learning Translation

Subtitling a course rather than a single video? E-learning translation covers the course text, the platform strings and the assessments alongside the subtitle files, with reading speed held low enough for learners taking notes.

Digital Marketing Translation

Subtitling a campaign that has landing pages and ads behind it? Digital marketing translation carries the same terminology from the video into the page, so the ad and the destination do not read as two different companies.

2,000+

Linguists

ISO 17100

Certified quality standard

1 hour

Free quote

150+

Languages Supported

What Else Do Clients Ask About Subtitling Services?


How much do subtitling services cost per video minute?

Subtitling starts at $5 per video minute, so a 20-minute video is quoted on 20 minutes. That covers the timed transcript where you do not have one, the spotting, the translation, the second-linguist revision and the finished subtitle file. What moves a quote above the floor is the work in the material: how much dialogue the video carries, the language pair, the output format and the deadline. Translating a subtitle file you already hold is quoted instead on word count at $0.10 per word. Send the video, or a representative five minutes of it, and an itemized quote comes back free within 1 hour.

Are subtitling services free, and what do the free options actually give you?

The free options are the automatic captions a platform generates on upload, and for some videos they are enough: one speaker, clean audio, an audience that already speaks the language, nothing riding on the file. They stop at three places. They do not translate into a language your audience reads, they time to the waveform rather than to the picture, and they take no account of reading speed or line length, so they fail any published specification. Where none of those three matters to you, use the free version and keep the money.

What is the difference between a subtitle and a caption?

A subtitle carries the dialogue for a viewer who can hear the audio but does not speak the language, usually translated and condensed to fit the screen. A caption carries the dialogue in the language it was spoken in, close to verbatim, for a viewer who cannot hear it, and it also reports the non-speech sound that carries meaning. One solves a language problem and the other solves a hearing problem, which is why they are not interchangeable: a caption file does nothing for a foreign-language audience, and a translated subtitle track leaves a deaf viewer without the sound cues. SDH is the file that does both jobs inside one subtitle track.

How do subtitles work?

A subtitle file is a numbered list of cues, each one holding a start time, an end time and one or two lines of text. The player reads that list and paints each line over the picture at the time it is told to, so nothing is written into the video and the viewer can switch subtitles off or change language. In an SRT the times are written hours:minutes:seconds with a comma before the milliseconds and an arrow between start and end; a WebVTT file opens with the word WEBVTT and uses a period instead. Setting the in and out points against the picture is called spotting, and it is aligned to shot changes, because a line crossing a cut reads as a mistake however good the translation is. To add subtitles to a video you upload that file alongside it and the player does the rest.

Are captions or subtitles legally required in the United States?

Captions are required in some circumstances and translated subtitles are not required by either of the main rules. Under 47 CFR 79.4 the FCC requires closed captions on full-length video programming delivered over the internet only where that programming was published or exhibited on television in the United States with captions, and the rule expressly excludes consumer-generated media and clips. Separately, Section 508 requires public-facing electronic content to meet WCAG 2.0 Level A and AA, and the captions criterion for prerecorded synchronized media sits at Level A. Neither asks for a second language. Check your own obligation with counsel rather than with a supplier; what we can tell you is which file each rule is describing.

How long does subtitling take?

Standard subtitling projects are delivered in 2 to 3 working days, with 24-hour express available when a release date is fixed. A feature-length film, a series, or one video going into several languages at once takes longer, and the exact date is confirmed in the quote before any work starts. Tell us the deadline in the first message rather than after the quote: a date named early is cheaper to meet than one discovered late.

What subtitle file formats do you deliver?

SRT, VTT, STL, TTML, SCC and DFXP, plus burned-in subtitles for platforms that strip a subtitle file out of the upload. SRT suits YouTube, Vimeo, social platforms and most web players; VTT is the HTML5 format; EBU-STL and SCC are broadcast formats; TTML and DFXP carry positioning and styling for streaming delivery. All of them are built from one set of timings, so a second format is a conversion rather than a new job, and every file is checked for cue numbering, timing, line length and reading speed before it is sent.

Can you subtitle a video if I have no script or transcript?

Yes, and that is the normal case. A subtitler transcribes the dialogue and times it against the picture first, then translates inside those cues; the timed transcript is part of the $5 per video minute, not an extra. If you do hold a script, a transcript or an existing subtitle file, send it: translation starts immediately, the existing time codes are kept, and the job is quoted on word count at $0.10 per word instead.

How accurate are YouTube’s automatic subtitles?

Good enough for in-platform search and a clear single speaker, and not good enough for a published video with an audience that depends on the text. Automatic systems mishear proper nouns and technical terms, lose grammar when a sentence is split across two cues, time to the audio rather than to the shot, and translate idiom literally. None of that is visible to the person who approved the file and all of it is visible to a viewer who speaks the language. We would not quote an accuracy percentage for any subtitle file, our own included, because no standard test produces one; what you can compare is the process behind it.

Do you provide SDH subtitles and closed captions?

Yes. We produce closed captions in the source language, SDH for deaf and hard-of-hearing audiences with speaker labels, sound effects and music cues, and translated SDH where an international audience needs both the language and the sound information. SDH is the file to order when the platform carries only one subtitle track. Say which of the three you need at the quote stage, because they are different work and the wrong one is no use to the audience it was bought for.

Do Netflix and other streaming platforms have their own subtitling requirements?

Yes, and they are published, which makes them easy to build to. Netflix’s English timed-text guide allows 42 characters per line and two lines at most, a reading speed up to 20 characters per second for adult programs and 17 for children’s, and a cue duration between five-sixths of a second and seven seconds. Other distributors and broadcasters publish their own numbers and their own delivery formats. Name the platform when the project is set up and the file is built to that specification from the start, rather than corrected against it afterwards.

Can you subtitle into several languages from one video?

Yes. One source video can produce subtitle files in as many of our 150+ languages as you need, and multilingual sets are built from a single set of timings so every language lines up with the same picture. A dedicated Project Manager holds terminology, speaker labels and formatting consistent across the set. Foreign language subtitles for international distribution are quoted per language per video minute, and ordering the set together costs less than ordering them one at a time. One video can go to international clients in eight markets off one spotting pass.

How do I choose between a subtitling company and a freelance subtitler?

On the shape of the job rather than on the price. A single video into one language, with no platform specification attached, is well handled by a good freelance subtitler. A company earns its place on the things one person cannot do alone: a second linguist revising the translation, several languages held consistent across one set of timings, multiple output formats, a fixed point of contact when a release date moves, and cover when somebody is ill. Ask either of them the five questions on this page — who makes the transcript, who revises and against what, which readability numbers they build to, how they test condensation, and which formats come from one spotting pass.

Can you caption a conference or a webinar as it happens?

No. We do not provide live captioning or live subtitling, and we will not put a captioner or a respeaker in your event. Everything on this page is work on a recording: you send a file, a subtitler spots, translates and condenses it, a second linguist revises it, and a subtitle file comes back. Text appearing on screen while somebody is still speaking is a different trade with different equipment, and a supplier who does both is doing two jobs rather than one. What we do cover is everything after the event. The replay, the edited highlights and the clips cut out of them get properly spotted subtitles in any of 150+ languages from $5 per video minute, which is the version that stays online and takes most of the views. Tell us the date before the event and the recording is scheduled for the day it lands.

Contact Us


Get in touch with our team for professional translation services. We're here to help you with all your translation needs.

Email
info@espressotranslations.com

Phone
+1 332 9006150

Espresso Translations Google Reviews rating
Espresso Translations Trustpilot rating
ISO 17100 certified translation services

Request a Quote Now

We’ll get back to you with a competitive quote.