Updating to latest #1
Open
Conversation
This file contains 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
* Update resampling logic to handle chunking properly * Cleanup logging * Optimize memory usage when resampling * Add filter to input prompt text * Correct timestamp filter logic for #170 * Filter out zero length segments - when calculating word timestamps - resolves #170 * Add method for async audio loading * Fix async load audio function * Fix tests * Fix tests * Fix tests * Revert timestamp filter changes * Temporarily remove xcpretty for tests * Check suspected test crash * Remove errant test case for japanese options * Add bigger range for early stopping test * Reset progress between runs * Fix progress resetting and improve example app transcription handling * Update tests * Minimize crash risk for early stop checks * Fix finalize text * Add source text to language label
…lity (#192) * Make additional initializers, functions, members public, for WKPro * Allows use of default internal functions & member accesses which have increased protections when imported * Initializers were Xcode generated: right click class name -> refactor -> generate memberwise initializers * memberwise initializer defaults to internal, mark as public. * Formatting --------- Co-authored-by: ZachNagengast <znagengast@gmail.com>
… models (#193) * Add initial mlpackage loading (if .mlmodelc not present) -- Does not modify model loading in OS WK. This is a hook to modify load path URLs. * Always load audio encoder last * Adjust timings to account for decoder<>encoder order swap * Add helper for mlpackage detection --------- Co-authored-by: ZachNagengast <znagengast@gmail.com>
* Fix start time logic for file loading and resampling * Add test file
As far as I can tell, these stored properties are not meant to be changed. Therefore, change them to be immutable. This change also makes these static properties concurrency-safe.
* Add VoiceActivityDetector base class Add base class to allow different VAD implementations * fix spaces
…hen chunkingStrategy is vad (#205)
* CI fetch depth 0 * VAD refactoring * Update logo * Add WhisperKitConfig * Open whisperkit methods * add missing @available --------- Co-authored-by: BlaiseMuhirwa <blaisemuhirwa3@gmail.com> Co-authored-by: ZachNagengast <znagengast@gmail.com>
* Add model support config fetching from model repo * Fix audio start index error handling Co-authored-by: 1amageek <1amageek@users.noreply.github.com> * Formatting * Fix CI + watchOS build - New github runner image does not include visionOS, so to prevent downloading for all platforms this will specify the platform from the test matrix * Fix typo * Use dispatch group for sync recommendedModels * Remove sync remote model fetching * Formatting and cleanup from review --------- Co-authored-by: 1amageek <1amageek@users.noreply.github.com>
* Release memory when transcribing single files Co-authored-by: keleftheriou <keleftheriou@users.noreply.github.com> * Add method to load from file into float array iteratively - Reduces peak memory by doing the array conversion while loading in chunks so the array copy size is lower - Previously copied the entire buffer which spiked the memory 2x * Fix leak * Use vad by default in examples * Fix vad thread issue * Fix unused warning * Revert change to early stop callback * Fix warnings - Optional cli commands are deprecated - @_disfavoredOverload required @available to prevent infinite loop * PR review - simplify early stop test logic Co-authored-by: Andrey Leonov <aleonov@gmail.com> * Cleanup from review --------- Co-authored-by: keleftheriou <keleftheriou@users.noreply.github.com> Co-authored-by: Andrey Leonov <aleonov@gmail.com>
Co-authored-by: Arda Atahan Ibis <ardaibis@gmail.com> Co-authored-by: ZachNagengast <znagengast@gmail.com>
* Fix xcconfig tracking * Add package.swift docs to readme * Fix edge case where framePosition does not align with actual frame count of AVAudioFile * Upgrade github runner macos version * Update remaining github runner versions * Use WERUtils to check vad accuracy * Reduce calls to frameposition * Fix xcode version for runner
* SegmentDiscovery callback * ModelState callback * FractionCompleted callback * TranscriptionPhaseCallback callback * Updates for review * Formatting * Remove remaining callback from init --------- Co-authored-by: ZachNagengast <znagengast@gmail.com>
* Use @Frozen for loglevel enum * Add public accessibility for loggingcallback and loglevel
# for free
to join this conversation on GitHub.
Already have an account?
# to comment
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.
No description provided.