---
title: "Comment by Camille Dubois on \"Conditioning a video model with audio: three reference clips, and where lip sync actually stops\" | Hashnode"
description: "Three reference clips is more than I expected. On mobile we mostly care about whether the audio survives the encode - does the model re-render the audio track,"
url: "https://hashnode.com/posts/h3-reference-audio-conditioning/6ae6f6c173e9f4171f4e133a/comment/6ae6f8e6849868cc5538463f"
---

[Discussion](https://hashnode.com/posts/h3-reference-audio-conditioning/6ae6f6c173e9f4171f4e133a)

Comment [Conditioning a video model with audio: three reference clips, and where lip sync actually stops](https://hashnode.com/posts/h3-reference-audio-conditioning/6ae6f6c173e9f4171f4e133a)

[Camille Dubois](https://hashnode.com/@camille-dubois)

[Nov 1](https://hashnode.com/posts/h3-reference-audio-conditioning/6ae6f6c173e9f4171f4e133a/comment/6ae6f8e6849868cc5538463f)

Three reference clips is more than I expected. On mobile we mostly care about whether the audio survives the encode - does the model re-render the audio track, or just condition on it and pass it through?
