The exported fitness clip is visually clean and playable, but it carries a persistent 'TRANSLATED BY RASK' watermark and the free-tier workflow had to be reviewed via screen recording rather than a native download.
What was measured
Output Quality & Export
Is the final exported video clean, downloadable, and production-ready, and are there watermarks or credit restrictions?
context, not decisivetransformation
A clean export and watermark-free download matter for usability, but they do not determine whether the tool can translate, clone voice, and lip-sync well. (3 of 3 judges)
What was given, what came back
Test input: Fitness instructor short (English → Hindi) · video · group: video-translation-voice-clone-lip-sync
Input — what we sent
Fitness instructor short (English → Hindi)
A short fitness/coaching video with energetic single-speaker English speech, used to test whether tools can translate into Hindi while preserving fast delivery, motivational tone, and original-face lip sync.
Why this input is hard
- · Fast, energetic speech transcription
- · Tone preservation for motivational fitness content
- · English-to-Hindi translation quality
- · Lip-sync accuracy on a moving face
- · End-to-end dubbing workflow automation
Output — unretouched
Also checked on this input — same tool, 2 other criteria
Lip Sync Accuracy✗ FailedAt matched timestamps the output stays frame-identical to the source, with no visible mouth or face regeneration, so the free-tier export does not perform lip sync.Translation Accuracy✗ FailedThe English→Hindi fitness export does not localize the visible on-screen text: the translated clip still shows English questions such as 'Which head of the shoulder is most active on this exercise?' and 'Can you name the three BCAAs?'.
Provenance
- Observation
- 54745121-4401-4bf1-b320-aae0810f7b34
- Evidence run
- ec1dd100-6af8-4f49-8f19-b78492702b14
- Study
- Translate Videos with Voice Cloning and Lip Sync Using AI
- Research task
- 86b96dfpf
- Tested at
- not recorded
- Source
- first-party
- Evidence state
- verified
- Proof shown
- input + output shown
- Cost / latency
- not captured
- Repeat run
- not captured
- Tester
- not captured
The last three rows are honest blanks, not placeholders — our capture has no field for them yet.
Query this
get_evidence({
tool: "rask-ai",
scenario: "video-translation-voice-clone-lip-sync"
})MCP · mcp.aidemos.com/api/mcp
Free with attribution.
Same input, same check — 4 other tools
measured on Output Quality & Export
Akool✗ FailedThe export visibly downscaled a 1080×1920 source to 404×720.Dubverse✓ WorkedIt exports a dubbed video with synced audio, offers MP4 and audio-only outputs, and the export time is described as moderate.ElevenLabs✗ FailedThe fitness dubbed export is not clean or production-ready because the output retains an ElevenLabs watermark.Synthesia✗ FailedThe visual compositing pass is unstable in the fitness run: the subject's body and white tank top turn a bright unnatural green for several seconds before returning to normal.
This evidence is published in
From the same study (page rebuilt from a later run)
Real inputs and real outputs, no retouching · every cell queryable via API & MCP · aidemos.com