Video AI for Developers
World-class APIs and infrastructure for video AI.
Trusted by leading video AI platforms
Kaiber logoVeed logoZight logoKapwing logoScenery logoCapsule logoScispace logoSwell AI logoAdventr logoRecast logoJupitrr logoRevid logo
Kaiber logoVeed logoZight logoKapwing logoScenery logoCapsule logoScispace logoSwell AI logoAdventr logoRecast logoJupitrr logoRevid logo
Meet our APIsHigh-quality, performant APIs built for your video AI use case.
Dubbing
Visual QA
Auto Crop
I want to ask you some very random questions
English
Audio waveformAudio waveformAudio waveformAudio waveformAudio waveform
Spanish
Audio waveformAudio waveformAudio waveformAudio waveformAudio waveform
Formal French
Audio waveformAudio waveformAudio waveformAudio waveformAudio waveform

sieve

/

dubbing

Localize content in any language
Preserve original tone and style
Broad & robust language support
3x realtime processing
Pick various voice & translation engines
Design, develop, deployBuilt for Developers
Sieve's platform allows you to customize existing applications to meet your specific cost, quality, and performance requirements or build your own.
Make tradeoffs between cost, quality and speedPick settings baked into apps that allow you to make direct tradeoffs for your use case.
USER INPUTS

visual_detail

Visual detail to process at

process_fps

Frame rate to process video at

speed_boost

Speeds up processing at the cost of quality

spoken_context

Use the transcript as context. Increases cost.

Build custom AI appsImport existing APIs and models like Python packages to build something new.

# Video dubbing in a few lines of code
transcribe = sieve.function.get("sieve/whisper")
translate = sieve.function.get("sieve/translate")
tts = sieve.function.get("sieve/xtts")
sync = sieve.function.get("sieve/video_retalking")

# Transcribe, translate, and generate audio
transcript = transcribe.run(source_video)
translated = translate.run(text, "eng", language)
speech = tts.run(source_audio, language, translated)

# Lipsync dubbed audio!
dubbed_video = sync.run(source_video, speech)

Performant infrastructureFast, scalable infrastructure that can handle video-scale.
Run at any scale
We built Sieve to automatically scale as your traffic increases with zero extra configuration.
Stop worrying about GPUs in the cloud
Package models with a simple Python decorator and deploy instantly.
Logs and metrics
A full-featured observability stack so you have full visibility of what’s happening under the hood.
Flexible, compute-based pricing
Pay only for what you use, by the second. Gain full control over your costs.
TestimonialsHear from our users
Kaiber uses Sieve for our production AI video gen workloads. Our end-to-end shipping speed would be notably slower without Sieve, as they handle so many common, yet complex requirements out of the box. In addition, their team is exceptionally communicative and have always worked quickly to address our feedback.
kaiber logoEric Gao, CTO
Zight uses Sieve to power all our production AI capabilities. Our team loves the fact that Sieve is one roof where all our video AI features can live under. It allows us to move at unparalleled speed from prototype → production.
kaiber logoPhin Hochart, Head of Product
Sieve strikes the right balance between production-ready solutions and a flexible platform to scale our GPU workloads. They also partnered with us to develop solutions and were hyper responsive to our needs.
kaiber logoJoseph Jorgensen, CTO
Built for your use case
Work with the experts
We're a team of machine learning & infrastructure experts from places like Berkeley, NVIDIA, Apple, Snapchat, Microsoft, and Scale AI. Work with us to support your custom use case and integrate Sieve into your existing pipeline.