Clean up eager processing and credit contributor#277
Open
peteonrails wants to merge 1 commit intomainfrom
Open
Clean up eager processing and credit contributor#277peteonrails wants to merge 1 commit intomainfrom
peteonrails wants to merge 1 commit intomainfrom
Conversation
Drop the cached eager_transcriber after each eager recording finishes (push-to-talk release, toggle stop, timeout, and external trigger paths). Previously it was only cleared on cancel, leaving a stale Arc<dyn Transcriber> between recordings that could hold model resources unnecessarily. Add Sami Jawhar to README contributors and website news for wiring eager input processing into the daemon event loop (#275). Co-authored-by: Sami Jawhar <sami@trajectorylabs.net>
Owner
Author
|
@sjawhar Will follow your PR up with this minor cleanup item and do a bunch of testing this weekend. It'll be in the tagged 6.5 release! |
Contributor
|
I think I'm noticing in my own testing of the original PR that there is some slight overlap between chunks sometimes. Planning to look into it soon |
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Follow-up to #275. Addresses review items and adds contributor credit.
eager_transcribercache after every eager recording stop (push-to-talk release, toggle stop, timeout, external trigger). Previously only cleared on cancel, which left a staleArc<dyn Transcriber>between recordings.Why this matters
With
on_demand_loading = trueor model switching, holding a reference to the transcriber between recordings prevents the model from being unloaded. Clearing the cache after each eager recording ensures model resources can be released when the daemon returns to idle.Test plan
cargo buildcompiles cleanlycargo test-- 547 tests passfinish_eager_recordingcall sites now clear the cacheCo-authored-by: Sami Jawhar sami@trajectorylabs.net