Skip to content
Permalink

Comparing changes

Choose two branches to see what’s changed or to start a new pull request. If you need to, you can also or learn more about diff comparisons.

Open a pull request

Create a new pull request by comparing changes across two branches. If you need to, you can also . Learn more about diff comparisons here.
base repository: itchat/Scribe
Failed to load repositories. Confirm that selected base ref is valid, then try again.
Loading
base: v2.1.0
Choose a base ref
...
head repository: itchat/Scribe
Failed to load repositories. Confirm that selected head ref is valid, then try again.
Loading
compare: v2.2.0
Choose a head ref
  • 2 commits
  • 12 files changed
  • 1 contributor

Commits on May 12, 2026

  1. Live Captions: add Qwen3-ASR 0.6B and 1.7B (MLX) engines 

    VAD-segmented streaming via Silero VAD + per-segment partial refresh,
since speech-swift's StreamingASR is pull-style and doesn't expose the
prefix-prompt knobs needed for a continuous rollback algorithm. Bumps
build VERSION to 2.2.0.
    @itchat
    itchat committed May 12, 2026
    Configuration menu
    Copy the full SHA
    d29a79c View commit details
    Browse the repository at this point in the history

Commits on May 28, 2026

  1. Qwen3-ASR: fix empty SRT for long videos, memory leak, and restart hang 

    - Offline: replace one-shot transcribe with VAD-segmented chunked
  transcription via Qwen3SegmentPlanner (force-split at 25s) so long
  videos produce full SRTs instead of near-empty ones.
- Streaming: remove early-return bug that blocked recognition after
  stop/start; cap MLX cache (512MB) and periodically clearCache to
  prevent memory growth over long Live Captions sessions.
- Tests: add Qwen3SegmentPlanner unit tests, streaming lifecycle tests,
  and extend offline E2E coverage.
    @itchat
    itchat committed May 28, 2026
    Configuration menu
    Copy the full SHA
    28e1e33 View commit details
    Browse the repository at this point in the history
Loading