chore: publishing server

Author: Eithan

.github/workflows/publish-client.yaml

@@ -39,7 +39,7 @@ jobs:
         with:
           images: eithan1231/gerald-assistant-client
           tags: |
-            type=raw,value={{date 'YYYY'}}-{{date 'MM'}}-{{date 'DD'}}-{{sha}}
+            type=raw,value={{date 'YYYY'}}-{{date 'MM'}}-{{date 'DD'}}-commit-{{sha}}
             type=raw,value=latest
 
 

.github/workflows/publish-server.yaml

@@ -0,0 +1,54 @@
+name: "Publish Docker Server image"
+
+on:
+  push:
+    branches:
+      - master
+
+jobs:
+  push_to_registry:
+    name: Push Docker image to Docker Hub
+
+    runs-on: ubuntu-latest
+
+    permissions:
+      packages: write
+      contents: read
+      attestations: write
+      id-token: write
+
+    steps:
+      - name: Check out the repo
+        uses: actions/checkout@v4
+
+      - name: Log in to Docker Hub
+        uses: docker/login-action@f4ef78c080cd8ba55a85445d5b36e214a81df20a
+        with:
+          username: ${{ secrets.DOCKER_USERNAME }}
+          password: ${{ secrets.DOCKER_PASSWORD }}
+
+      - name: Set up QEMU
+        uses: docker/setup-qemu-action@v3
+      
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@v3
+
+      - name: Docker meta
+        id: meta
+        uses: docker/metadata-action@v4
+        with:
+          images: eithan1231/gerald-assistant-server
+          tags: |
+            type=raw,value={{date 'YYYY'}}-{{date 'MM'}}-{{date 'DD'}}-commit-{{sha}}
+            type=raw,value=latest
+
+
+      - name: Build and push Docker image
+        id: push
+        uses: docker/build-push-action@3b5e8027fcad23fda98b2e3ac259d8d67585f671
+        with:
+          context: server
+          platforms: linux/amd64,linux/arm64
+          push: true
+          tags: ${{ steps.meta.outputs.tags }}
+          labels: ${{ steps.meta.outputs.labels }}

server/docker-compose.yaml

@@ -0,0 +1,57 @@
+version: "3.8"
+
+services:
+  server:
+    container_name: ga-server
+    image: eithan1231/gerald-assistant-server
+    restart: unless-stopped
+    environment:
+      # Listen words. Can be used for changing the name of your agent, configured to
+      # "jeff,jeffery,gerald" by default.
+      #
+      # Uncommon or rare names might be harder for the speach-to-text engine to pick up.
+      # Gerald is often picked up as "joe", which is not ideal, but hence the "Jeff"
+      - LISTEN_WORDS=jeff,jeffery,gerald
+
+      # Your OpenAI key geerated by OpenAI. This is required.
+      - OPENAI_KEY=SECRET_OPENAI_KEY
+
+      # mimic3, used for text-to-speach.
+      - ENDPOINT_TTS=http://mimic3:59125/api/tts
+
+      # faster-whisper-server, used for speach-to-text.
+      - ENDPOINT_TRANSCRIBE=http://faster-whisper-server:8000/v1/audio/transcriptions
+
+    volumes:
+      - /dev/snd:/dev/snd
+
+  faster-whisper-server:
+    container_name: faster-whisper-server
+    # To run model on GPU, utilise the tag "latest-cuda", for cpu use "latest-cpu".
+    # Most models will run sufficiently on CPU.
+    image: fedirz/faster-whisper-server:latest-cpu
+    restart: always
+    volumes:
+      # Mounting this volume will vastly improve startup times, and will avoid
+      # the model being re-downloaded
+      - ./cache/faster-whisper/huggingface:/root/.cache/huggingface
+    environment:
+      # faster-whisper models for speach-to-text. I have found the base-en model to be a great
+      # compromise between speed and performance. Hallucinations for "okay okay okay", or "please
+      # subscribe to xyz" are not entirely uncommon, but rare enough to not be impactful.
+      #
+      # 789MB - Systran/faster-distil-whisper-medium.en
+      # 322MB - Systran/faster-distil-whisper-small.en
+      # 145MB - Systran/faster-whisper-base.en
+      # 75MB - Systran/faster-whisper-tiny.en
+      #
+      # More can be found here:
+      # https://huggingface.co/Systran
+      - WHISPER_MODEL=Systran/faster-whisper-base.en
+
+  mimic3:
+    container_name: mimic3
+    image: mycroftai/mimic3
+    restart: always
+    volumes:
+      - ./cache/mimic3:/home/mimic3/.local/share/mycroft/mimic3

server/src/text-to-speech.ts

@@ -31,6 +31,12 @@ export const createTextToSpeech = async (
   const response = await fetch(`${endpoint}?${params.toString()}`);
 
   if (response.status !== 200) {
+    console.log(`[createTextToSpeech] Failed to TTS ${response.status}`);
+
+    return {
+      success: false,
+      message: "Failed to convert text to speech",
+    };
   }
 
   const responseContent: Buffer[] = [];