Support local-only mode (gpu)

README.md

@@ -94,6 +94,7 @@ services:
       - PGID=1000
       - TZ=Etc/UTC
       - WHISPER_MODEL=tiny-int8
+      - LOCAL_ONLY= #optional
       - WHISPER_BEAM=1 #optional
       - WHISPER_LANG=en #optional
     volumes:
@@ -112,6 +113,7 @@ docker run -d \
   -e PGID=1000 \
   -e TZ=Etc/UTC \
   -e WHISPER_MODEL=tiny-int8 \
+  -e LOCAL_ONLY= `#optional` \
   -e WHISPER_BEAM=1 `#optional` \
   -e WHISPER_LANG=en `#optional` \
   -p 10300:10300 \
@@ -131,6 +133,7 @@ Containers are configured using parameters passed at runtime (such as those abov
 | `-e PGID=1000` | for GroupID - see below for explanation |
 | `-e TZ=Etc/UTC` | specify a timezone to use, see this [list](https://en.wikipedia.org/wiki/List_of_tz_database_time_zones#List). |
 | `-e WHISPER_MODEL=tiny-int8` | Whisper model that will be used for transcription. From `tiny`, `base`, `small` and `medium`, all with `-int8` compressed variants |
+| `-e LOCAL_ONLY=` | If set to `true`, or any other value, the container will not attempt to download models from HuggingFace and will only use locally-provided models. |
 | `-e WHISPER_BEAM=1` | Number of candidates to consider simultaneously during transcription. |
 | `-e WHISPER_LANG=en` | Language that you will speak to the add-on. |
 | `-v /config` | Local path for Whisper config files. |
@@ -298,6 +301,7 @@ Once registered you can define the dockerfile to use with `-f Dockerfile.aarch64
 
 ## Versions
 
+* **10.08.25:** - Add support for local-only mode.
 * **05.12.24:** - Build from Github releases rather than Pypi.
 * **18.07.24:** - Rebase to Ubuntu Noble.
 * **19.05.24:** - Bump CUDA to 12 on GPU branch.

jenkins-vars.yml

@@ -6,8 +6,6 @@ external_type: github_stable
 release_type: prerelease
 release_tag: gpu
 ls_branch: gpu
-image_sbom: true
-image_provenance: true
 repo_vars:
   - EXT_GIT_BRANCH = 'master'
   - EXT_USER = 'rhasspy'

readme-vars.yml

@@ -13,7 +13,8 @@ available_architectures:
 development_versions: true
 development_versions_items:
   - {tag: "latest", desc: "Stable releases"}
-  - {tag: "gpu", desc: "Releases with Nvidia GPU support"}
+  - {tag: "gpu", desc: "Releases with Nvidia GPU support (amd64 only)"}
+  - {tag: "gpu-legacy", desc: "Legacy releases with Nvidia GPU support for pre-Turing cards (amd64 only)"}
 # container parameters
 common_param_env_vars_enabled: true
 param_container_name: "{{ project_name }}"
@@ -29,6 +30,7 @@ param_ports:
 # optional container parameters
 opt_param_usage_include_env: true
 opt_param_env_vars:
+  - {env_var: "LOCAL_ONLY", env_value: "", desc: "If set to `true`, or any other value, the container will not attempt to download models from HuggingFace and will only use locally-provided models."}
   - {env_var: "WHISPER_BEAM", env_value: "1", desc: "Number of candidates to consider simultaneously during transcription."}
   - {env_var: "WHISPER_LANG", env_value: "en", desc: "Language that you will speak to the add-on."}
 readonly_supported: true
@@ -83,6 +85,8 @@ init_diagram: |
   "faster-whisper:gpu" <- Base Images
 # changelog
 changelogs:
+  - {date: "20.08.25:", desc: "Add gpu-legacy branch for pre-Turing cards."}
+  - {date: "10.08.25:", desc: "Add support for local-only mode."}
   - {date: "05.12.24:", desc: "Build from Github releases rather than Pypi."}
   - {date: "18.07.24:", desc: "Rebase to Ubuntu Noble."}
   - {date: "19.05.24:", desc: "Bump CUDA to 12 on GPU branch."}

root/etc/s6-overlay/s6-rc.d/svc-whisper/run

@@ -12,4 +12,5 @@ exec \
         --beam-size "${WHISPER_BEAM:-1}" \
         --language "${WHISPER_LANG:-en}" \
         --data-dir /config \
-        --download-dir /config
+        --download-dir /config \
+        ${LOCAL_ONLY:+--local-files-only}