COMPLETE GUIDE

Video Translation: The Complete Guide (2026)

FAMILIAR RESEARCHUPDATED ALL ARTICLES
READ THIS INEnglishEspañolPortuguês中文日本語한국어FrançaisDeutschالعربيةBahasa IndonesiaРусскийItalianoTiếng ViệtไทยNederlandsΕλληνικάУкраїнська粵語NorskCatalàHrvatskiLietuviųSlovenčinaSvenskaБългарскиСрпскиБеларускаяDanskҚазақшаLatviešu

ABSTRACT

There are four ways to translate a video: subtitles, voice-over in a stranger's voice, voice-only dubbing in the speaker's own voice, and voice + lip-sync dubbing, where the performance is translated: lip-sync and everything past it. What each method preserves, what it costs per minute, and how to choose, whether you translate a video to English, to Spanish, or a whole catalog into 30 languages.

The four ways to translate a video

WHAT SURVIVES, BY METHOD · PRICE PER FINISHED MINUTE, PER LANGUAGE
MethodThe wordsThe voiceThe faceThe sceneTypical price
SubtitlesRead, not heardKept (original audio)KeptKeptIncluded with every Familiar dub · professional translation about $10/min
Voice-over (the classic dub)Spoken by a strangerReplacedMouths the original languageDepends on the studio mix$70 to $150 full human, translation plus voice · voice packages alone $39 to $94 · from $128 premium studio
Voice-only dubbingSpokenThe speaker's ownStill mouths the originalOften degraded (measured in Sec 02)$3 (ElevenLabs checkout quote, Aug 5, 2026), audio only
Voice + lip-sync dubbingSpokenThe speaker's ownLip-synced to the new language, up to 4KMusic, effects, ambience kept; true 5.1 or stereo mixFamiliar: one render, by Studio contract

YOUTUBE AUTO-DUBBING · VOICE-OVER

  • YouTube auto-dubbing and voice-over both put another voice on the speaker.

VOICE-ONLY DUBBING

  • Voice-only dubbing, often searched as "voice cloning," keeps the voice; the mouth still speaks the old language.

Voice + lip-sync dubbingAI LIP SYNC

The AI lip sync people search for, extended to the whole face. More: AI Video Dubbing, Explained.
HeyGenHeyGenElevenLabsElevenLabsHeyGen + ElevenLabsONE-SHOT TOOLS · NO CONTROL

One shot today means giving up the rest.

The one-shot tools today give up the rest: the translation cannot be edited, the music and the room drop out, the voice drifts, and the lips fail on a hand over the face, a microphone, blur, bokeh, a side view or several people in frame. No spatial reasoning, no understanding of the world, no control. ElevenLabs' current Dubbing v2 edits only through an Enterprise API, and its full editor is the older model, in maintenance mode; HeyGen stitches a lip-sync over a dub like that, and the lip-sync breaks on the ordinary shots of any episode. Familiar is one shot and the quality: translation first, reviewed; then the voice, the lips and the scene together, through the hard shots.

54.7%

FEWER TRANSLATION ERRORS · FAMILIAR VS ELEVENLABS DUBBING V2

270%

ELEVENLABS MORE BACKGROUND-SOUND ERROR · THE LAUGHTER, THE MUSIC, THE AMBIENCE

+27.8%

FAMILIAR CLOSER TO THE REAL SPEAKER

FAMILIAR · ONE SHOT AND THE QUALITYFamiliar vs Flawless TrueSync The ElevenLabs study The demos

What quality means and how it is measured

The honest test is paired: the same source video to each system.

Two paired studies against ElevenLabs, never pooled, at /benchmark/elevenlabs, judged on:

MEANINGVOICE CLONINGSCENE NOISE BEDTRANSIENT TIMING
FIG. 01 · REVIEW-FLAGGED SPOKEN-OUTPUT MISTAKES · STABLE DUBBING V1 · 418 PAIRED OUTPUTS · 11 LANGUAGES
ElevenLabs258
Familiar132
  • Same study: Familiar sounds 27.8% more like the real speaker (0.4605 vs 0.3604).
  • All 11 languages favored Familiar.
ElevenLabsElevenLabsElevenLabsTHEIR DOCS, READ OCTOBER 30, 2026

Editing the translation means their old model.

On ElevenLabs' current Dubbing v2, in alpha, editing the translation and regenerating the audio through the API is for their Enterprise plans only. Their full editor, Dubbing Studio, is "only available via the v1 model" and "in maintenance mode", in their words. Dubbing v1 is the model our 418-pair study measured, and its output is mono.

+27.8%

FAMILIAR CLOSER TO THE REAL SPEAKER · ALL 11 LANGUAGES

261%

ELEVENLABS MORE BACKGROUND ERROR · THE MUSIC, THE AMBIENCE

+243.7%

FAMILIAR ON LAUGHTER AND REACTION SHAPE

ELEVENLABS DUBBING V1 · 418 PAIRED OUTPUTS · 11 LANGUAGES · AUGUST 5, 2026Read the study Their docs

The Dubbing v2 (Alpha) study, 111 paired outputs, shows the same shape:

270%

ElevenLabs produced 270% more background-sound error: the laughter, the music, the ambience.

11.92 VS 3.22 DB · DUBBING V2 (ALPHA)
54.7%

Familiar made 54.7% fewer important spoken translation errors.

29 VS 64 · DUBBING V2 (ALPHA)
+28.2%

Familiar scored higher on laughter and reaction shape correlation.

0.960 VS 0.749 · DUBBING V2 (ALPHA)

What translation quality is achievable

August 2026: Familiar against expert professional translators, scored blind against expert reference edits.

FIG. 02 · TRANSLATION QUALITY VS THE FIRST-PASS EXPERT PROFESSIONAL · DASHED LINE = THE PROFESSIONAL'S 100%
French100.0%
Chinese98.2%
Hindi103.3%
Indonesian99.6%
FRENCH, CHINESE, HINDI, INDONESIAN (FIRST PASS FOR EXPERT PROFESSIONAL TRANSLATORS)

Control is the other half, and it is where Familiar's Multimodal Localization Decision Model (MLDM), the first foundation model for multimodal translation, a new paradigm and the first of its kind, shows its decisions:

DO NOT TRANSLATE
A Do Not Translate list keeps names and catchphrases exactly as the speaker says them.
JOKES AND LOADED LINES
Jokes and loaded lines come back as several renderings, each with its reasoning, for your creative directors to pick.
REVIEW SCREEN
Every video pauses at a review screen: edit any line before render. Every decision is recorded with the job and can be overridden.
IN THE WORKS
Keeping a chant or a yell in the original language, picking a different face for a line, and choosing which text in the frame to translate.
THE FIRST FOUNDATION MODEL FOR MULTIMODAL TRANSLATION

Cost per minute

FIG. 03 · PRICE PER FINISHED MINUTE, PER TARGET LANGUAGE · ELEVENLABS = $3/MIN CHECKOUT QUOTE · AUG 5, 2026 · RANGES AT MIDPOINT
ElevenLabs$3 audio only
Subtitles, pro~$10 text only
Human voice$39–94
Full human dub$70–150
Studio dubfrom $128
None of these rates includes lip-sync. The assistive route to it takes a team of VFX artists and weeks per project; the one-shot tools today are fast and give up the rest. Familiar is one shot and the quality: one render from the MLDM carries the translation, the speaker's own voice, the scene audio, and the lip-sync.
The subtitle figure is the median professional rate across the 11 study languages.

PICTURE

Up to 4K

MIX

True 5.1 or stereo

SUBTITLES

.srt · .vtt

With every dub, cut to Netflix's Timed Text Style Guide: two lines, 42 characters a line, up to 17 a second.

ACCESS

Studio contract

Sized to your catalog; 30-day money back on signed annual contracts.

Full human dubbing, translation plus voice, runs $70 to $150 per finished minute.

That is $4,200 to $9,000 for a 60-minute video in one language, so even top productions dub only a few languages and smaller productions, shows and vertical dramas don't dub at all.

Access: /pricing; the full rate survey: How Much Does Dubbing Cost in 2026?

Choosing by content type

TALKING-HEAD VIDEOS
WHAT DECIDES ITThe face carries the message, so voice + lip-sync dubbing fits.
PODCASTS AND INTERVIEWS
WHAT DECIDES ITMultiple speakers must each keep their own voice.
MOVIES AND TV
WHAT DECIDES ITMusic, effects, and reaction timing decide whether a dub is watchable; Sec 02 measures those. Studio playbooks: Why the Dub Feels Off Abroad, and How Studios Fix It and What Is the Best AI Dubbing for Movies?
STREAMING CATALOGS
WHAT DECIDES ITCoverage is the constraint: dub on upload, review per language, release each language when it clears. How Streaming Platforms Close the Local-Language Gap.
MICRO-DRAMAS
WHAT DECIDES ITDaily episodes in the actors' own voices, with names and catchphrases held by a series-level Do Not Translate list. Dubbing Micro-Dramas on a Daily Release Schedule.
SIMULTANEOUS LAUNCHES
WHAT DECIDES ITOne request into every target language, each market released on its own approval. Multi-Country Launches: Dubbing Every Market at Once.
COURSES AND LECTURES
WHAT DECIDES ITTerminology is the risk: put technical terms on the Do Not Translate list and check them at the review screen.
SERMONS
WHAT DECIDES ITA weekly recording reaching a multilingual congregation, in the speaker's own voice rather than an interpreter's.
LIVESTREAMS
WHAT DECIDES ITOnly live dubbing fits here; subtitles and recorded dubs arrive too late. At Familiar, live dubbing is in the works.
ADS AND PRODUCT VIDEO
WHAT DECIDES ITShort runtimes make a complete dub a small line item, and the founder's or spokesperson's real voice carries the trust.
FIG. 04 · THE 100 BIGGEST YOUTUBE CHANNELS, BY LANGUAGE
  • Not in English64%
  • In English36%

Familiar's 30 languages run any-to-any: a Spanish or Korean channel can translate a video to English, the biggest audience unlock.

REFERENCES

  1. [1]Netflix, Timed Text Style Guide: Subtitle Templates, read October 30, 2026
  2. [2]Familiar vs ElevenLabs benchmark: full results, intervals, and listening examples
  3. [3]AI vs Human Translation: Tested Against Professionals

Multimodal Translation

Familiar Labs builds generationally better dubbing with a Multimodal Localization Decision Model (MLDM), the first foundation model for multimodal translation: one model decides what to localize across text, audio and visual, and shows you the choices. Your voice and lips, carried into new languages. They're a team of 8 with PhDs, backed by Samsung, RedBird, Gold House. Our mission

See the full benchmark →Talk with the team