Update to Opencast 15.0

.github/workflows/cron.yml

@@ -26,24 +26,24 @@ jobs:
             opencast_version: 'master'
             push_major: 'false'
           - day_of_week: '1'
-            opencast_version: '14.7'
+            opencast_version: '15.0'
             push_major: 'true'
           - day_of_week: '2'
-            opencast_version: '14.6'
+            opencast_version: '15.0'
             push_major: 'false'
           - day_of_week: '3'
-            opencast_version: '14.5'
+            opencast_version: '15.0'
             push_major: 'false'
 
           # legacy
           - day_of_week: '4'
-            opencast_version: '13.13'
+            opencast_version: '14.7'
             push_major: 'true'
           - day_of_week: '5'
-            opencast_version: '13.12'
+            opencast_version: '14.6'
             push_major: 'false'
           - day_of_week: '6'
-            opencast_version: '13.11'
+            opencast_version: '14.5'
             push_major: 'false'
 
     steps:

Dockerfile

@@ -27,20 +27,29 @@ RUN apk add --no-cache \
  && mv ff* /usr/local/bin
 
 
-FROM docker.io/eclipse-temurin:11-jdk AS build-su-exec
+FROM docker.io/eclipse-temurin:17-jdk AS build-whisper-cpp
+ARG WHISPER_CPP_VERSION="master"
 RUN apt-get update \
  && apt-get install -y --no-install-recommends \
+      g++ \
       gcc \
       git \
       libc-dev \
       make
-RUN git clone https://github.com/ncopa/su-exec.git /tmp/su-exec \
- && cd /tmp/su-exec \
- && make \
- && cp su-exec /usr/local/sbin
+RUN mkdir -p /tmp/whisper.cpp
+WORKDIR /tmp/whisper.cpp
+RUN git clone https://github.com/ggerganov/whisper.cpp.git . \
+ && git checkout "$WHISPER_CPP_VERSION"
+RUN make -j \
+ && sed -i 's#models_path=.*$#models_path=/usr/share/whisper.cpp/models/#' models/download-ggml-model.sh
+RUN mkdir -p out \
+ && mv main out/whisper.cpp \
+ && mv quantize out/whisper.cpp-quantize \
+ && mv server out/whisper.cpp-server \
+ && mv models/download-ggml-model.sh out/whisper.cpp-model-download
 
 
-FROM --platform=${BUILDPLATFORM} docker.io/maven:3-eclipse-temurin-11 AS build-opencast
+FROM --platform=${BUILDPLATFORM} docker.io/maven:3-eclipse-temurin-17 AS build-opencast
 
 ARG OPENCAST_REPO="https://github.com/opencast/opencast.git"
 ARG OPENCAST_VERSION="develop"
@@ -81,8 +90,8 @@ ARG OPENCAST_DISTRIBUTION
 RUN tar -xzf build/opencast-dist-${OPENCAST_DISTRIBUTION}-*.tar.gz --strip 1 -C "${OPENCAST_HOME}"
 
 
-FROM docker.io/eclipse-temurin:11-jdk
-LABEL org.opencontainers.image.base.name="docker.io/eclipse-temurin:11-jdk"
+FROM docker.io/eclipse-temurin:17-jdk
+LABEL org.opencontainers.image.base.name="docker.io/eclipse-temurin:17-jdk"
 
 ENV OPENCAST_HOME="/opencast" \
     OPENCAST_DATA="/data" \
@@ -91,16 +100,17 @@ ENV OPENCAST_HOME="/opencast" \
     OPENCAST_GROUP="opencast" \
     OPENCAST_UHOME="/home/opencast" \
     OPENCAST_UID="800" \
-    OPENCAST_GID="800"
+    OPENCAST_GID="800" \
+    WHISPER_CPP_MODELS="/usr/share/whisper.cpp/models"
 ENV OPENCAST_CONFIG="${OPENCAST_HOME}/etc" \
     OPENCAST_SCRIPTS="${OPENCAST_HOME}/docker/scripts" \
     OPENCAST_STAGE_BASE_HOME="${OPENCAST_HOME}/docker/stage/base" \
     OPENCAST_STAGE_OUT_HOME="${OPENCAST_HOME}/docker/stage/out"
 
 RUN groupadd --system -g "${OPENCAST_GID}" "${OPENCAST_GROUP}" \
  && useradd --system -M -N -g "${OPENCAST_GROUP}" -d "${OPENCAST_UHOME}" -u "${OPENCAST_UID}" "${OPENCAST_USER}" \
- && mkdir -p "${OPENCAST_DATA}" "${OPENCAST_UHOME}" \
- && chown -R "${OPENCAST_USER}:${OPENCAST_GROUP}" "${OPENCAST_DATA}" "${OPENCAST_UHOME}"
+ && mkdir -p "${OPENCAST_DATA}" "${OPENCAST_UHOME}" "${WHISPER_CPP_MODELS}" \
+ && chown -R "${OPENCAST_USER}:${OPENCAST_GROUP}" "${OPENCAST_DATA}" "${OPENCAST_UHOME}" "${WHISPER_CPP_MODELS}"
 
 RUN apt-get update \
  && apt-get install -y --no-install-recommends \
@@ -111,31 +121,20 @@ RUN apt-get update \
       fonts-freefont-ttf \
       fonts-liberation \
       fonts-linuxlibertine \
-      hunspell \
-      hunspell-en-au \
-      hunspell-en-ca \
-      hunspell-en-gb \
-      hunspell-en-us \
-      hunspell-en-za \
+      gosu \
       inotify-tools \
       jq \
       netcat-openbsd \
       openssl \
-      python3 \
-      python3-pip \
       rsync \
-      sox \
-      synfig \
       tesseract-ocr \
       tesseract-ocr-eng \
       tzdata \
- && pip install \
-      vosk-cli \
  && rm -rf /var/lib/apt/lists/*
 
-COPY --from=build-su-exec   /usr/local/sbin/su-exec  /usr/local/sbin/
-COPY --from=build-ffmpeg    /usr/local/bin/ff*       /usr/local/bin/
-COPY --from=build-opencast  "${OPENCAST_HOME}"       "${OPENCAST_HOME}"
+COPY --from=build-ffmpeg       /usr/local/bin/ff*      /usr/local/bin/
+COPY --from=build-whisper-cpp  /tmp/whisper.cpp/out/*  /usr/local/bin/
+COPY --from=build-opencast     "${OPENCAST_HOME}"      "${OPENCAST_HOME}"
 COPY rootfs /
 
 ARG OPENCAST_REPO="https://github.com/opencast/opencast.git"

Dockerfile-build

@@ -27,21 +27,30 @@ RUN apk add --no-cache \
  && mv ff* /usr/local/bin
 
 
-FROM docker.io/maven:3-eclipse-temurin-11 AS build-su-exec
+FROM docker.io/eclipse-temurin:17-jdk AS build-whisper-cpp
+ARG WHISPER_CPP_VERSION="master"
 RUN apt-get update \
  && apt-get install -y --no-install-recommends \
+      g++ \
       gcc \
       git \
       libc-dev \
       make
-RUN git clone https://github.com/ncopa/su-exec.git /tmp/su-exec \
- && cd /tmp/su-exec \
- && make \
- && cp su-exec /usr/local/sbin
+RUN mkdir -p /tmp/whisper.cpp
+WORKDIR /tmp/whisper.cpp
+RUN git clone https://github.com/ggerganov/whisper.cpp.git . \
+ && git checkout "$WHISPER_CPP_VERSION"
+RUN make -j \
+ && sed -i 's#models_path=.*$#models_path=/usr/share/whisper.cpp/models/#' models/download-ggml-model.sh
+RUN mkdir -p out \
+ && mv main out/whisper.cpp \
+ && mv quantize out/whisper.cpp-quantize \
+ && mv server out/whisper.cpp-server \
+ && mv models/download-ggml-model.sh out/whisper.cpp-model-download
 
 
-FROM docker.io/maven:3-eclipse-temurin-11
-LABEL org.opencontainers.image.base.name="docker.io/maven:3-eclipse-temurin-11"
+FROM docker.io/maven:3-eclipse-temurin-17
+LABEL org.opencontainers.image.base.name="docker.io/maven:3-eclipse-temurin-17"
 
 ARG OPENCAST_REPO="https://github.com/opencast/opencast.git"
 ARG OPENCAST_VERSION="develop"
@@ -74,7 +83,8 @@ ENV OPENCAST_REPO="${OPENCAST_REPO}" \
     OPENCAST_UHOME="/home/opencast" \
     OPENCAST_UID="800" \
     OPENCAST_GID="800" \
-    OPENCAST_BUILD_ASSETS="/docker"
+    OPENCAST_BUILD_ASSETS="/docker" \
+    WHISPER_CPP_MODELS="/usr/share/whisper.cpp/models"
 ENV OPENCAST_CONFIG="${OPENCAST_HOME}/etc" \
     OPENCAST_SCRIPTS="${OPENCAST_HOME}/docker/scripts" \
     OPENCAST_STAGE_BASE_HOME="${OPENCAST_HOME}/docker/stage/base" \
@@ -111,40 +121,29 @@ RUN apt-get update \
       fonts-freefont-ttf \
       fonts-liberation \
       fonts-linuxlibertine \
-      hunspell \
-      hunspell-en-au \
-      hunspell-en-ca \
-      hunspell-en-gb \
-      hunspell-en-us \
-      hunspell-en-za \
+      gosu \
       inotify-tools \
       jq \
       netcat-openbsd \
       openssl \
-      python3 \
-      python3-pip \
       rsync \
-      sox \
-      synfig \
       tesseract-ocr \
       tesseract-ocr-eng \
       tzdata \
- && pip install \
-      vosk-cli \
   \
  && sudo ln -s /usr/local/openjdk*/bin/* /usr/local/bin/ \
   \
  && groupadd --system -g "${OPENCAST_GID}" "${OPENCAST_GROUP}" \
  && useradd --system -M -N -g "${OPENCAST_GROUP}" -d "${OPENCAST_UHOME}" -u "${OPENCAST_UID}" "${OPENCAST_USER}" \
- && mkdir -p "${OPENCAST_SRC}" "${OPENCAST_HOME}" "${OPENCAST_UHOME}" "${OPENCAST_DATA}" "${OPENCAST_BUILD_ASSETS}" \
- && chown -R "${OPENCAST_USER}:${OPENCAST_GROUP}" "${OPENCAST_HOME}" "${OPENCAST_UHOME}" "${OPENCAST_DATA}" \
+ && mkdir -p "${OPENCAST_SRC}" "${OPENCAST_HOME}" "${OPENCAST_UHOME}" "${OPENCAST_DATA}" "${OPENCAST_BUILD_ASSETS}" "${WHISPER_CPP_MODELS}" \
+ && chown -R "${OPENCAST_USER}:${OPENCAST_GROUP}" "${OPENCAST_HOME}" "${OPENCAST_UHOME}" "${OPENCAST_DATA}" "${WHISPER_CPP_MODELS}" \
  && echo "opencast-builder ALL = NOPASSWD: ALL" > /etc/sudoers.d/opencast-builder \
   \
  && cd / \
  && rm -rf /tmp/* /var/lib/apt/lists/*
 
-COPY --from=build-su-exec  /usr/local/sbin/su-exec  /usr/local/sbin/
-COPY --from=build-ffmpeg   /usr/local/bin/ff*       /usr/local/bin/
+COPY --from=build-ffmpeg       /usr/local/bin/ff*      /usr/local/bin/
+COPY --from=build-whisper-cpp  /tmp/whisper.cpp/out/*  /usr/local/bin/
 COPY rootfs       "${OPENCAST_BUILD_ASSETS}/"
 COPY rootfs-build /
 

Makefile

@@ -14,11 +14,12 @@
 
 # user configurable variables
 
-VERSION           ?= $(shell cat VERSION)
-VERSION_MAJOR     ?= $(shell cat VERSION_MAJOR)
-OPENCAST_REPO     ?= https://github.com/opencast/opencast.git
-OPENCAST_VERSION  ?= $(shell cat VERSION_OPENCAST)
-FFMPEG_VERSION    ?= $(shell cat VERSION_FFMPEG)
+VERSION             ?= $(shell cat VERSION)
+VERSION_MAJOR       ?= $(shell cat VERSION_MAJOR)
+OPENCAST_REPO       ?= https://github.com/opencast/opencast.git
+OPENCAST_VERSION    ?= $(shell cat VERSION_OPENCAST)
+FFMPEG_VERSION      ?= $(shell cat VERSION_FFMPEG)
+WHISPER_CPP_VERSION ?= $(shell cat VERSION_WHISPER_CPP)
 
 IMAGE_REGISTRY           ?= quay.io/opencast
 IMAGE_TAGS               ?= latest $(VERSION) $(VERSION_MAJOR)
@@ -58,6 +59,7 @@ build-%:
 		--build-arg OPENCAST_VERSION="$(OPENCAST_VERSION)" \
 		--build-arg OPENCAST_DISTRIBUTION="$*" \
 		--build-arg FFMPEG_VERSION="$(FFMPEG_VERSION)" \
+		--build-arg WHISPER_CPP_VERSION="$(WHISPER_CPP_VERSION)" \
 		--build-arg BUILD_DATE="$(BUILD_DATE)" \
 		--build-arg GIT_COMMIT="$(GIT_COMMIT)" \
 		--build-arg VERSION="$(VERSION)" \
@@ -74,6 +76,7 @@ build-build:
 		--build-arg OPENCAST_REPO="$(OPENCAST_REPO)" \
 		--build-arg OPENCAST_VERSION="$(OPENCAST_VERSION)" \
 		--build-arg FFMPEG_VERSION="$(FFMPEG_VERSION)" \
+		--build-arg WHISPER_CPP_VERSION="$(WHISPER_CPP_VERSION)" \
 		--build-arg BUILD_DATE="$(BUILD_DATE)" \
 		--build-arg GIT_COMMIT="$(GIT_COMMIT)" \
 		--build-arg VERSION="$(VERSION)" \

README.md

@@ -17,6 +17,7 @@
     -   [Database](#database)
         -   [H2](#h2)
         -   [MariaDB and PostgreSQL](#mariadb-and-postgresql)
+    -   [whisper.cpp](#whisper.cpp)
     -   [Miscellaneous](#miscellaneous)
 -   [Data](#data)
 -   [Languages](#languages)
@@ -50,6 +51,8 @@ If you want to build the images yourself, there is a `Makefile` with the necessa
      The name of the Git branch, tag or commit hash to check out. Defaults to the content of the `VERSION_OPENCAST` file.
 -   `FFMPEG_VERSION`<br>
      The version of the Opencast FFmpeg build. Defaults to the content of the `VERSION_FFMPEG` file.
+-   `WHISPER_CPP_VERSION`<br>
+     The version of whisper.cpp. Defaults to the content of the `VERSION_WHISPER_CPP` file.
 -   `IMAGE_REGISTRY`<br>
      The first part of the image name. It defaults to `quay.io/opencast` and will be extended by the name of the Opencast distribution.
 -   `IMAGE_TAG`<br>
@@ -75,7 +78,7 @@ In the `./docker-compose` directory there are also compose files for more produc
 
 ## Images
 
-Opencast comes in different distributions. For each of the official distributions, there is a specific container image. Each version is tagged. For example, the full image name containing the `admin` distribution at version `14.7` is `quay.io/opencast/admin:14.7`. Leaving the version out will install the latest one.
+Opencast comes in different distributions. For each of the official distributions, there is a specific container image. Each version is tagged. For example, the full image name containing the `admin` distribution at version `15.0` is `quay.io/opencast/admin:15.0`. Leaving the version out will install the latest one.
 
 ### `allinone`
 
@@ -200,6 +203,11 @@ There are no additional environment variables you can set if you are using the H
 -   `ORG_OPENCASTPROJECT_DB_JDBC_PASS` **Required**<br>
     Password of the database user. You may alternatively set `ORG_OPENCASTPROJECT_DB_JDBC_PASS_FILE` to the location of a file within the container that contains the password.
 
+### whisper.cpp
+
+-   `WHISPER_CPP_DOWNLOAD_MODEL` Optional<br>
+    Download the specified whisper.cpp model. Note that models are downloaded in parallel to the startup of Opencast. Not waiting for model downloads greatly improves the Opencast startup time. However, transcription operations may fail if models are not available yet. The default is an empty string, i.e. no model is downloaded.
+
 ### Miscellaneous
 
 -   `TIMEZONE` Optional<br>
@@ -211,7 +219,7 @@ The data directory is located at `/data`. Use [Docker Volumes](https://docs.dock
 
 ## Languages
 
-Opencast makes use of [Tesseract](https://github.com/tesseract-ocr/tesseract) to recognize text in videos (ORC) and [Hunspell](https://hunspell.github.io/) to identify spelling mistakes. Both need additional files namely [traningsdata](https://github.com/tesseract-ocr/tessdata) and [dictionaries](http://download.services.openoffice.org/contrib/dictionaries) respectively. These images come with files for the English language. If you need other or more languages you can use Docker Volumes to mount them in the appropriate directories `/usr/share/tessdata` and `/usr/share/hunspell`.
+Opencast makes use of [Tesseract](https://github.com/tesseract-ocr/tesseract) to recognize text in videos (ORC). For this, [training data files](https://github.com/tesseract-ocr/tessdata) are needed. These images come with files for the English language. If you need other or more languages you can extend these images or use Docker Volumes to mount them in the appropriate directories `/usr/share/tessdata`.
 
 ## References
 

VERSION

@@ -1 +1 @@
-14.7
+15.0

VERSION_FFMPEG

@@ -1 +1 @@
-6.0
+6.1

VERSION_MAJOR

@@ -1 +1 @@
-14
+15

VERSION_OPENCAST

@@ -1 +1 @@
-14.7
+15.0

VERSION_WHISPER_CPP

@@ -0,0 +1 @@
+v1.5.2

docker-compose/docker-compose.allinone.h2+pyca.yml

@@ -32,7 +32,7 @@ services:
       - opensearch:/usr/share/opensearch/data
 
   opencast:
-    image: quay.io/opencast/allinone:14.7
+    image: quay.io/opencast/allinone:15.0
     environment:
       ORG_OPENCASTPROJECT_SERVER_URL: http://opencast:8080
       ORG_OPENCASTPROJECT_DOWNLOAD_URL: http://localhost:8080/static

docker-compose/docker-compose.allinone.h2.yml

@@ -31,7 +31,7 @@ services:
       - opensearch:/usr/share/opensearch/data
 
   opencast:
-    image: quay.io/opencast/allinone:14.7
+    image: quay.io/opencast/allinone:15.0
     environment:
       ORG_OPENCASTPROJECT_SERVER_URL: http://localhost:8080
       ORG_OPENCASTPROJECT_DOWNLOAD_URL: http://localhost:8080/static

docker-compose/docker-compose.allinone.mariadb.yml

@@ -43,7 +43,7 @@ services:
       - opensearch:/usr/share/opensearch/data
 
   opencast:
-    image: quay.io/opencast/allinone:14.7
+    image: quay.io/opencast/allinone:15.0
     environment:
       ORG_OPENCASTPROJECT_SERVER_URL: http://localhost:8080
       ORG_OPENCASTPROJECT_DOWNLOAD_URL: http://localhost:8080/static

docker-compose/docker-compose.allinone.postgres.yml

@@ -41,7 +41,7 @@ services:
       - opensearch:/usr/share/opensearch/data
 
   opencast:
-    image: quay.io/opencast/allinone:14.7
+    image: quay.io/opencast/allinone:15.0
     environment:
       ORG_OPENCASTPROJECT_SERVER_URL: http://localhost:8080
       ORG_OPENCASTPROJECT_DOWNLOAD_URL: http://localhost:8080/static

docker-compose/docker-compose.build.yml

@@ -31,7 +31,7 @@ services:
       - opensearch:/usr/share/opensearch/data
 
   opencast:
-    image: quay.io/opencast/build:14.7
+    image: quay.io/opencast/build:15.0
     tty: true
     stdin_open: true
     environment:

docker-compose/docker-compose.multiserver.build.yml

@@ -43,7 +43,7 @@ services:
       - opensearch:/usr/share/opensearch/data
 
   opencast-admin:
-    image: quay.io/opencast/build:14.7
+    image: quay.io/opencast/build:15.0
     tty: true
     stdin_open: true
     environment:
@@ -71,7 +71,7 @@ services:
       - "${OPENCAST_SRC}:/usr/src/opencast"
 
   opencast-presentation:
-    image: quay.io/opencast/build:14.7
+    image: quay.io/opencast/build:15.0
     tty: true
     stdin_open: true
     environment:
@@ -98,7 +98,7 @@ services:
       - "${OPENCAST_SRC}:/usr/src/opencast"
 
   opencast-worker:
-    image: quay.io/opencast/build:14.7
+    image: quay.io/opencast/build:15.0
     tty: true
     stdin_open: true
     environment: