START HERE
The video workspace is for existing footage. Manage source media and timing on the desktop, use configured cloud recognition, a dedicated translation model and speech synthesis, then preview and export with local media tools. Review captions, translation and audio separately and reprocess only the necessary scope.
Turn a recognition draft into reviewed captions
Confirm source and target languages and the estimate after import. Recognition supplies timing; edit text, segmentation and retained sections. AI correction and segmentation produce proposals for review, not established facts. Recognition is currently planned from source duration; retained edit ranges do not limit recognition to those ranges.
Consider terminology and readability together
Select an accessible glossary version and provide context, approved equivalents and tone. Adapt bilingual captions to aspect ratio, font and line limits. Translucent backgrounds aid contrast, while short-cue merge suggestions can reduce fragmentation; inspect proposals before applying them.
Arrange translated speech on the timeline
Choose a voice with explicit target-language support and preview it before generating many cues. New projects default to 1× synthesis. Timeline fitting uses actual audio duration and has limits; excess duration requires reviewed shortening or editing. Mix original and translated audio separately; completion order does not determine playback order.
Local media work with cloud language processing
Parsing, waveforms, editing and export use desktop media tools. Recognition, translation and speech need configured cloud services and account allowance; this is not fully offline processing. Check subtitle format, aspect ratio, audio and a final sample before delivery. Each project currently starts from one source asset.
A practical comparison
| Stage | Processing | Confirm |
|---|---|---|
| Parsing / editing / export | Local desktop | Media, storage and output |
| Recognition | Configured recognition service | Language, hotwords and duration |
| Translation / AI proposals | Dedicated translation model | Context, terms and review |
| Dubbing | Speech service and local mixing | Voice, cost and measured timing |
Before you sign off
- Install a supported current Windows or macOS client.
- Ask the administrator to confirm access and recognition, translation and speech resources.
- Test an independent short asset before confirming full-video cost and delivery.
Frequently asked questions
Why is video production missing after sign-in?
The entry depends on administrator enablement. A compatible desktop client and account configuration are also required; check version and access first.
Is everything processed on my computer?
No. Parsing, editing and export are local, while recognition, translation, AI processing and speech use the configured services.
Does retaining five minutes mean only five minutes are recognised?
No. Recognition is planned from source duration; retained ranges control editing and export. Prepare a separate short asset for a trial and check the estimate.
Scope and related product information
Scope: current Wali desktop clients for Windows and macOS. Video subtitle production requires administrator enablement and configured recognition, dedicated translation and speech resources. Availability depends on the account and version. A mobile browser is not the local video export workspace.
Examples in this guide illustrate a review method, not a customer case or a performance result. Product availability changes with version and account configuration.
Video subtitle production and translated dubbing | Wali desktop →Current clients and installation guidance →