Master the audible companion, not just the picture - #310
Merged
Merged
Conversation
The loudness pass reached the video track and stopped there. The audible companion is encoded from the raw narration rather than from the finished video, so it never inherited that mastering and kept the level the synthesised voice came back at. That is the half of the path most listeners actually get. A music or radio player asks /api/ads/stream for kind=audio, because a video creative has nowhere to go in a player showing album art, and the companion is the only file it ever plays. So the spot that was brought up to streaming level was the one nobody in an audio slot hears. Measured on a companion the worker rendered after the video fix shipped: -28.5 dB mean before, -19.6 after, peak -3.0. Same target and ceiling as the video track, and a test pins them together: two levels for one spot would mean the advert changed loudness when a listener moved between a video slot and an audio one. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
ThreatCrush Security Scan49 finding(s) HIGH/CRITICAL: 2 | MEDIUM: 32 | LOW: 15
Snippets are redacted; ThreatCrush never prints matched credential material. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The loudness pass in #307 reached the video track and stopped there. The audible companion is encoded from the raw narration rather than from the finished video, so it never inherited that mastering and kept the level the synthesised voice came back at.
That is the half of the path most listeners actually get. A music or radio player asks
/api/ads/streamforkind=audio, because a video creative has nowhere to go in a player showing album art, and the companion is the only file it ever plays. So the spot that was brought up to streaming level was the one nobody in an audio slot hears.Measured on a companion the worker rendered after #307 shipped, so this is not staleness:
Same target and ceiling as the video track. A test pins the two together: two levels for one spot would mean the advert changed loudness when a listener moved between a video slot and an audio one.
Found while checking whether a nixamp live would play a correctly levelled advert. It would not: nixamp's slot is active and filling, and it asks for
kind=audio.Note for after the merge: this fixes what the worker renders next. The 183 already published spots keep their current audio until they are re-rendered.
🤖 Generated with Claude Code