feat(whisper-cpp): Convert to Purego and add VAD (#6087)

* fix(ci): Avoid matching wrong backend with the same prefix

Signed-off-by: Richard Palethorpe <io@richiejp.com>

* chore(whisper): Use Purego and enable VAD

This replaces the Whisper CGO bindings with our own Purego based module
to make compilation easier.

In addition this allows VAD models to be loaded by Whisper. There is not
much benefit now except that the same backend can be used for VAD and
transcription. Depending on upstream we may also be able to use GPU for
VAD in the future, but presently it is disabled.

Signed-off-by: Richard Palethorpe <io@richiejp.com>

---------

Signed-off-by: Richard Palethorpe <io@richiejp.com>
Co-authored-by: Ettore Di Giacinto <mudler@users.noreply.github.com>
This commit is contained in:
Richard Palethorpe
2025-08-28 16:25:18 +01:00
committed by GitHub
parent ead00a28b9
commit e6ebfd3ba1
13 changed files with 424 additions and 198 deletions

View File

@@ -20730,7 +20730,8 @@
- filename: nomic-embed-text-v1.5.f16.gguf
uri: https://huggingface.co/mradermacher/nomic-embed-text-v1.5-GGUF/resolve/main/nomic-embed-text-v1.5.f16.gguf
sha256: af8cb9e4ca0bf19eb54d08c612fdf325059264abbbd2c619527e5d2dda8de655
- name: "silero-vad"
- &silero
name: "silero-vad"
icon: https://github.com/snakers4/silero-models/raw/master/files/silero_logo.jpg
url: github:mudler/LocalAI/gallery/virtual.yaml@master
urls:
@@ -20750,6 +20751,22 @@
- filename: silero-vad.onnx
uri: https://huggingface.co/onnx-community/silero-vad/resolve/main/onnx/model.onnx
sha256: a4a068cd6cf1ea8355b84327595838ca748ec29a25bc91fc82e6c299ccdc5808
- !!merge <<: *silero
name: "silero-vad-ggml"
urls:
- https://github.com/snakers4/silero-vad
- https://github.com/ggml-org/whisper.cpp
- https://huggingface.co/ggml-org/whisper-vad
overrides:
backend: whisper-vad
parameters:
model: ggml-silero-v5.1.2.bin
options:
- "vad_only"
files:
- filename: ggml-silero-v5.1.2.bin
uri: https://huggingface.co/ggml-org/whisper-vad/resolve/main/ggml-silero-v5.1.2.bin
sha256: 29940d98d42b91fbd05ce489f3ecf7c72f0a42f027e4875919a28fb4c04ea2cf
- &bark
name: "bark-cpp"
icon: https://avatars.githubusercontent.com/u/99442120