Make whisper-asr independently scalable (HPA, multi-node)

Shipped

July 8, 2026 at 12:38 AM UTC

Author

Kamo

Commit

15e6150

Remove the single-node pin and switch to a per-pod model cache so pods can schedule on any node, and add an HPA (min 1 / max 4, CPU 75%, deliberate scale-up given slow model cold-starts). whisper-asr is the shared transcription backend for both meeting recordings and VOIP recordings/voicemails.