SpeakVid — a privacy-first, open, all-in-one video translation & dubbing tool. Transcribe, translate, dub and subtitle in one workflow. Local-first: your footage stays on your machine.
Every feature comes from real creative workflows — no bloated gimmicks.
We're our own toughest users. We use it daily before shipping it to you.
Models keep improving. We keep integrating the latest capabilities.
A good tool should hand the choices to creators: run local engines for free offline work, or bring your own API key for maximum quality. No lock-in, no dark patterns — every step transparent and under your control.
In local-engine mode, your video never leaves your computer — free and offline. In cloud mode, only the necessary audio segments are uploaded, then deleted after processing.
Recognition, translation and dubbing all run on industry-leading models with multi-level fallback, so output stays reliable.
Every stage is configurable and swappable. Bring your own key for the best services — never locked into a single vendor.
Recognition, translation and voice models improve every week. We keep integrating the latest capabilities.
Fully automated — and every step works standalone.
Local video or audio files, or paste a link and let us download it.
Local Whisper or cloud ASR, 24 languages, auto-segmented.
DeepSeek, Kimi, Qwen and more — 34 target languages.
300+ voices. Export video, subtitle files or MP3.
Every stage connects to industry-leading capabilities — and can be swapped anytime.
YouTube and TikTok creators generate multi-language dubbed versions in one click — reach viewers in more countries without relying on subtitle plugins.
Content exportOnline educators produce multi-language subtitles and voiceover at scale — sell one course to the whole world at near-zero marginal cost.
EducationTurn product videos into multiple languages fast, slashing content costs for overseas listings and social ads.
E-commerceProduce multi-language subtitle drafts in minutes — AI handles recognition and translation, humans just review and polish.
Media localizationLocal engines are completely free with unlimited use. Cloud engines run on a free monthly quota (60 minutes for new users) — or bring your own API key and pay only for what you use.
By default, local-engine mode keeps your video on your machine. Only when you choose a cloud engine are the necessary audio segments uploaded, then deleted after processing. We never keep your footage.
Speech recognition supports 24 languages (including Cantonese); translation supports 34 target languages. Common video and audio formats are accepted, and you can paste a video link for automatic download.
Yes. Transcribe, translate, dub and subtitle replacement each work standalone — pick only what you need.
Dubbed video (video + new audio track), subtitle files (SRT and more), standalone MP3 audio, and editable subtitle scripts for further editing.