Velorn v0.3.28
Velorn v0.3.28
Velorn v0.3.28 fixes Qwen ASR caption transcription for Music Video workflows on installations using newer versions of TTS-Audio-Suite.
Fixed
- Qwen ASR transcription compatibility. Velorn now detects whether the installed Qwen3-TTS Engine expects
model_variantor the oldermodel_sizeinput and queues the compatible caption workflow automatically. - Music Video timed lyrics. The same compatibility handling applies when preparing song lyrics and when transcribing a timeline for captions.
- Resilient schema lookup. If ComfyUI cannot report the node schema, Velorn safely provides both recognized input aliases so compatible TTS-Audio-Suite versions can continue.
Downloads
Windows Installer: standard Windows install experience for most usersWindows Portable: no-install Windows build for quick testing or portable useMac (Apple Silicon): for M1, M2, M3, M4, and newer MacsMac (Intel): for older Intel-based MacsLinux AppImage: portable Linux buildLinux deb: Debian/Ubuntu package
Ignore the auto-generated source-code archives unless you plan to build Velorn from source.