MAJORFailure Reference

OpenAI Sora Lip Sync Failure — Pre-Generation Risk Reference

Technical Classification

Audio-Visual Lip Sync & Phoneme Alignment Failure

Sora 2's integrated audio is a significant advance, but lip sync still fails on dialogue-driven prompts. The phoneme-to-viseme alignment is learned statistically rather than ground-truth — the audio says "absolutely" but the mouth shape sequence reads more like "okay-okay." Drift compounds over clip length: hands look fine at 1s and badly misaligned by 5s. Output is unusable for narration, commercials, or any final-frame dialogue work.

How to identify this failure

  • Mouth motion lagging audio by 100–400ms
  • Wrong viseme shape for the audible phoneme
  • Mouth opening during silent passages
  • Mouth closed during continued speech
  • Sync drift accumulating across clip length

Real generation examples

Prompt used

"Person saying 'absolutely fantastic' enthusiastically to camera"

Failure observed @ 0:01

Viseme misaligned with plosive at 0:01; mouth motion lagged audio ~200ms throughout

Prompt used

"Narrator delivering a 5-second monologue about coffee"

Failure observed @ 0:04

Sync drifted progressively; by 0:04 mouth motion was a half-word behind audio

Documentation strength

If you need to escalate

HIGH — OpenAI support refunds documented lip-sync drift on Sora 2 through the standard credit-refund flow. Provide Generation ID, prompt, audio track, and screen recording.

AVA is a pre-purchase prevention tool, not a post-purchase recovery tool. Platforms generally do not guarantee credit refunds for output-quality failures; goodwill credits are at each platform's discretion. The strength rating reflects how well-formed your support ticket can be, not a promised outcome.

Prevention + documentation steps

  1. 01

    Score your prompt before you generate

    Run your prompt through AVA's pre-flight scoring against the Audio-Visual Lip Sync & Phoneme Alignment Failure pattern. Green light = generate. Yellow/red = rewrite using the suggested fix before you commit credits.

  2. 02

    Capture Generation ID + timestamp if it failed anyway

    Find the Generation ID in the URL or share link. Note the exact time when the Audio-Visual Lip Sync & Phoneme Alignment Failure first appears (e.g. "failure first visible at 1.2s"). Timestamped evidence is significantly stronger than a general complaint.

  3. 03

    Use the correct technical term in your support ticket

    Describe this failure as "Audio-Visual Lip Sync & Phoneme Alignment Failure". This term maps to a recognised internal workflow in the support system and routes the ticket to the right team.

  4. 04

    Submit via the correct support channel

    Runway has no direct email intake. Pro+ plan: open the in-app AI Assistant (help widget bottom-right of app.runwayml.com), describe the failure with the technical term, attach evidence. Free/Standard plan: human support isn't available — your channel is Discord #community-help with @On Call - Moderators.

Frequently asked questions

Does Sora 2 refund credits for lip sync failures?

Yes. Submit the Generation ID with a screen recording showing the audio waveform alongside the mouth motion. OpenAI billing recognises lip sync drift as a known Sora 2 limitation on dialogue prompts.

Why does Sora fail at lip sync if audio is integrated?

Sora 2 generates audio and video jointly via a shared latent space, but the alignment is learned statistically — not enforced via ground-truth phoneme-to-viseme mapping. Longer clips accumulate misalignment as temporal consistency decays.

How do I get usable lip sync from Sora?

Keep dialogue clips ≤3 seconds. Use simple sentences with strong consonants (plosives like P/B/T anchor better than vowel-heavy phrases). Re-time audio in post if the output is close-but-not-tight. AVA flags lip-sync risk prompts before submission.

Score your prompt

Score your prompt against this failure mode in 30 seconds

Paste your prompt and the platform you intend to use. AVA returns a red/yellow/green score against this specific failure mode plus a concrete rewrite if the risk is high.

AVA Pro · founders' round

$50 for 6 months of unlimited scoring across all failure modes + personal failure-history dashboard. Locks in $13/mo grandfathered after.

Claim $50 founders

Related failures across models

If you’re seeing this failure, you may also encounter these on other models:

Pick a different tool for Sora failures

Some prompt shapes will keep failing on Sora. Routing those shots to a different vendor is the cheapest fix.