Treat streamed chunks as appends, stop Livewire from remorphing the Alpine tree, and hydrate from HTTP while a run is active so segment cards appear as they arrive.
Persist accumulated verbose_json on the recording and broadcast only transcript/whisper deltas so the show page can render language, segments, word timestamps, and confidence while a run is in progress.
Consume faster-whisper SSE instead of blocking on one JSON response, persist growing text and progress over Reverb, and drop the fake percent heartbeat.
Implement a new method for calculating elapsed transcription time in the Recording model. Enhance the TranscriptionService to handle local Whisper requests, including asynchronous processing and progress reporting. Update the recordings index view to display transcription progress percentage. Add unit tests for the new transcription service functionality.
Ship FrankenPHP Compose services with Reverb WebSockets for real-time transcription status, skip re-uploading identical audio via content hash, and format disk usage without requiring intl.
Batch uploads were only storing files as pending without dispatching Whisper jobs; queue them immediately, add a bulk pending action, and show human-readable status labels.
Raise queue retry_after above the job timeout, recover orphaned runs, allow stop/restart with any engine, and keep finished transcripts searchable in the recordings list.
ben