Merge branch 'master' into feat/streaming

.github/workflows/docker-publish.yml

@@ -35,9 +35,10 @@ jobs:
         with:
           images: ${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}
           tags: |
-            type=semver,pattern={{version}}
+            type=semver,pattern=v{{version}}
-            type=semver,pattern={{major}}.{{minor}}
+            type=semver,pattern=v{{major}}.{{minor}}
-            type=semver,pattern={{major}}
+            type=semver,pattern=v{{major}}
+            type=raw,value=latest
 
       # Extract metadata for CPU image
       - name: Extract metadata (tags, labels) for CPU Docker
@@ -48,9 +49,10 @@ jobs:
           flavor: |
             suffix=-cpu
           tags: |
-            type=semver,pattern={{version}}
+            type=semver,pattern=v{{version}}-cpu
-            type=semver,pattern={{major}}.{{minor}}
+            type=semver,pattern=v{{major}}.{{minor}}-cpu
-            type=semver,pattern={{major}}
+            type=semver,pattern=v{{major}}-cpu
+            type=raw,value=latest-cpu
 
       # Build and push GPU version
       - name: Build and push GPU Docker image
@@ -83,9 +85,10 @@ jobs:
           flavor: |
             suffix=-ui
           tags: |
-            type=semver,pattern={{version}}
+            type=semver,pattern=v{{version}}-ui
-            type=semver,pattern={{major}}.{{minor}}
+            type=semver,pattern=v{{major}}.{{minor}}-ui
-            type=semver,pattern={{major}}
+            type=semver,pattern=v{{major}}-ui
+            type=raw,value=latest-ui
 
       # Build and push UI version
       - name: Build and push UI Docker image

README.md

@@ -10,7 +10,7 @@
 Dockerized FastAPI wrapper for [Kokoro-82M](https://huggingface.co/hexgrad/Kokoro-82M) text-to-speech model
 - OpenAI-compatible Speech endpoint, with voice combination functionality
 - NVIDIA GPU accelerated inference (or CPU) option
-- very fast generation time (~30x real time factor via 4060Ti)
+- very fast generation time (~30x real time speed via 4060Ti)
 - automatic chunking/stitching for long texts
 - streaming support w/ variable chunking to control latency
 - simple audio generation web ui utility

docker-compose.cpu.yml

@@ -26,9 +26,11 @@ services:
       start_period: 1s
 
   kokoro-tts:
-    build:
+    image: ghcr.io/remsky/kokoro-fastapi:latest-cpu
-      context: .
+    # Uncomment below to build from source instead of using the released image
-      dockerfile: Dockerfile.cpu
+    # build:
+    #   context: .
+    #   dockerfile: Dockerfile.cpu
     volumes:
       - ./api/src:/app/api/src
       - ./Kokoro-82M:/app/Kokoro-82M
@@ -50,8 +52,10 @@ services:
 
   # Gradio UI service [Comment out everything below if you don't need it]
   gradio-ui:
-    build:
+    image: ghcr.io/remsky/kokoro-fastapi:latest-ui
-      context: ./ui
+    # Uncomment below to build from source instead of using the released image
+    # build:
+    #   context: ./ui
     ports:
       - "7860:7860"
     volumes:

docker-compose.yml

@@ -56,8 +56,10 @@ services:
 
   # Gradio UI service [Comment out everything below if you don't need it]
   gradio-ui:
-    build:
+    image: ghcr.io/remsky/kokoro-fastapi:latest-ui
-      context: ./ui
+    # Uncomment below to build from source instead of using the released image
+    # build:
+    #   context: ./ui
     ports:
       - "7860:7860"
     volumes: