feat: Optimized configuration

.github/workflows/ci-cd.yaml

@@ -1,20 +1,18 @@
-name: CI/CD Pipeline
-
-on:
-  push:
-    branches: [ main ]
-  # pull_request:
-  #   branches: [ main ]
-
+name: Build container
 env:
-  VERSION: 0.0.2
+  VERSION: 0.0.3
   REGISTRY: https://harbor.bwgdi.com
   REGISTRY_NAME: harbor.bwgdi.com
-  IMAGE_NAME: voxcpmtts 
-
+  REGISTRY_PATH: library
+  DOCKER_NAME: voxcpmtts
+on:
+  push:
+    branches:
+      - main
+  workflow_dispatch:
 jobs:
   build-docker:
-    runs-on: ubuntu-latest
+    runs-on: builder-ubuntu-latest
     steps:
       - name: Checkout
         uses: actions/checkout@v3
@@ -32,4 +30,4 @@ jobs:
           context: .
           file: ./Dockerfile
           push: true
-          tags: ${{ env.REGISTRY_NAME }}/library/${{env.IMAGE_NAME}}:${{ env.VERSION }}
+          tags: ${{ env.REGISTRY_NAME }}/${{ env.REGISTRY_PATH }}/${{ env.DOCKER_NAME }}:${{ env.VERSION }}

.gitignore

@@ -1,4 +1,42 @@
 launch.json
-__pycache__
 voxcpm.egg-info
-.DS_Store
+.DS_Store
+
+# Python-generated files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# Distribution / packaging
+build/
+dist/
+wheels/
+*.egg-info/
+
+# Unit test / coverage reports
+.pytest_cache/
+.coverage
+htmlcov/
+coverage.xml
+
+# Logs
+*.log
+log/*.log
+
+# Virtual environments
+.venv/
+venv/
+env/
+
+# IDE settings
+.vscode/
+.idea/
+
+# OS generated files
+.DS_Store
+
+# Generated files
+*.wav
+*.pdf
+
+*.lock

Dockerfile

@@ -7,6 +7,11 @@ RUN apt-get update && apt-get install -y \
 WORKDIR /app
 COPY api_concurrent.py requirements.txt ./
 RUN pip install -r requirements.txt
+ENV VOXCPM_MODEL_ID="/models/VoxCPM1.5/" \
+    VOXCPM_CPU_WORKERS="2" \
+    VOXCPM_UVICORN_WORKERS="1" \
+    MAX_GPU_CONCURRENT="1"
+
 EXPOSE 5000
 CMD [ "python", "./api_concurrent.py" ]
 

README_BW.md

@@ -6,11 +6,18 @@ https://github.com/BoardWare-Genius/VoxCPM
 
 | Version | Date       | Summary                         |
 |---------|------------|---------------------------------|
+| 0.0.3   | 2026-02-09 | Optimized configuration & Model support |
 | 0.0.2   | 2026-01-21 | Supports streaming               |
 | 0.0.1   | 2026-01-20 | Initial version                 |
 
 ### 🔄 Version Details
 
+#### 🆕 0.0.3 – *2026-02-09*
+- ✅ **Configuration & Deployment**
+  - Supports configuring model path via `VOXCPM_MODEL_ID`
+  - Supports configuring CPU workers via `VOXCPM_CPU_WORKERS`
+  - Supports configuring Uvicorn workers via `VOXCPM_UVICORN_WORKERS`
+
 #### 🆕 0.0.2 – *2026-01-21*
 
 - ✅ **Core Features**
@@ -27,9 +34,20 @@ https://github.com/BoardWare-Genius/VoxCPM
 
 # Start
 ```bash
-docker pull harbor.bwgdi.com/library/voxcpmtts:0.0.2
+docker pull harbor.bwgdi.com/library/voxcpmtts:0.0.3
 
-docker run -d --restart always -p 5001:5000 --gpus '"device=0"' --mount type=bind,source=/Workspace/NAS11/model/Voice/VoxCPM,target=/models harbor.bwgdi.com/library/voxcpmtts:0.0.2
+# Run with custom configuration
+# -e VOXCPM_MODEL_ID: Path to the model directory inside container
+# -e VOXCPM_CPU_WORKERS: Number of threads for CPU-bound tasks
+# -e VOXCPM_UVICORN_WORKERS: Number of uvicorn workers
+# -e MAX_GPU_CONCURRENT: Max concurrent GPU tasks
+docker run -d --restart always -p 5001:5000 --gpus '"device=0"' \
+  -e VOXCPM_MODEL_ID="/models/VoxCPM1.5/" \
+  -e VOXCPM_CPU_WORKERS="2" \
+  -e VOXCPM_UVICORN_WORKERS="1" \
+  -e MAX_GPU_CONCURRENT="1" \
+  --mount type=bind,source=/Workspace/NAS11/model/Voice/VoxCPM,target=/models \
+  harbor.bwgdi.com/library/voxcpmtts:0.0.3
 ```
 
 # Usage

api_concurrent.py

@@ -25,7 +25,7 @@ class VoxCPMDemo:
         self.device = "cuda" if torch.cuda.is_available() else "cpu"
         print(f"🚀 Running on device: {self.device}")
         self.voxcpm_model: Optional[voxcpm.VoxCPM] = None
-        self.default_local_model_dir = "/models/VoxCPM1.5/"
+        self.default_local_model_dir = os.environ.get("VOXCPM_MODEL_ID", "/models/VoxCPM1.5/")
 
     def _resolve_model_dir(self) -> str:
         if os.path.isdir(self.default_local_model_dir):
@@ -136,7 +136,8 @@ MAX_GPU_CONCURRENT = int(os.environ.get("MAX_GPU_CONCURRENT", "1"))
 gpu_semaphore = asyncio.Semaphore(MAX_GPU_CONCURRENT)
 
 # Use a thread pool for running blocking (CPU/GPU-bound) code.
-executor = ThreadPoolExecutor(max_workers=2)
+MAX_CPU_WORKERS = int(os.environ.get("VOXCPM_CPU_WORKERS", "2"))
+executor = ThreadPoolExecutor(max_workers=MAX_CPU_WORKERS)
 
 @app.on_event("shutdown")
 def shutdown_event():
@@ -159,6 +160,7 @@ async def generate_tts(
     retry_badcase_ratio_threshold: float = Form(6.0),
     prompt_wav: Optional[UploadFile] = None,
 ):
+
     try:
         prompt_path = None
         if prompt_wav:
@@ -266,4 +268,5 @@ async def root():
     return {"message": "VoxCPM API running 🚀", "endpoints": ["/generate_tts"]}
 
 if __name__ == "__main__":
-    uvicorn.run("api_concurrent:app", host="0.0.0.0", port=5000, workers=4)
+    uvicorn_workers = int(os.environ.get("VOXCPM_UVICORN_WORKERS", "1"))
+    uvicorn.run("api_concurrent:app", host="0.0.0.0", port=5000, workers=uvicorn_workers)

pyproject.toml

@@ -47,8 +47,7 @@ dependencies = [
     "funasr",
     "spaces",
     "argbind",
-    "safetensors"
-
+    "safetensors",
 ]
 
 [project.optional-dependencies]