The English dub is polluted by untranslated Romanized-Hindi caption text; 'KAR RAHE HAIN' and 'MAIN' appear in the output, and 'MAIN' is described as a fabricated caption that was not present in the source frame.

⚠ Struggled🧾 artifact-verifiedinput + output shownTest date not recordedVEED
What was measured
Translation Accuracy

Is the translated output semantically correct, and does it preserve technical terms, tone, and meaning?

decisive for this rankingtransformation

If the translation changes meaning, tone, or technical terms, the tool fails the core job of translating the video. (3 of 3 judges)

What was given, what came back

Test input: Hindi vlog-style talking head (Hindi → English) · video · group: video-translation-voice-clone-lip-sync
Input — what we sent
Input file 1 — as supplied
Input file 2 — as supplied
Hindi vlog-style talking head (Hindi → English)

A casual Hindi talking-head/vlog-style video used to test English dubbing on informal speech, Hinglish-like phrasing, speaker personality preservation, and lip sync under natural creator-style delivery.

Why this input is hard
  • · Informal Hindi speech transcription
  • · Hindi-to-English translation of casual phrasing
  • · Voice cloning for conversational creator tone
  • · Handling of slang/Hinglish-style content
  • · Lip-sync stability during expressive speech
Output — unretouched
Provenance
Observation
1e1f8faf-3b07-4aa3-ab63-88da22e098c5
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: "veed-io",
  scenario: "video-translation-voice-clone-lip-sync"
})
MCP · mcp.aidemos.com/api/mcp
Free with attribution.
Same input, same check — 8 other tools
measured on Translation Accuracy
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