add docker

Dockerfile

@@ -0,0 +1,19 @@
+FROM docker.io/nvidia/cuda:12.0.0-cudnn8-runtime-ubuntu22.04
+
+RUN apt-get update && \
+    apt-get install -y ffmpeg python3 python3-pip && \
+    apt-get clean && \
+    rm -rf /var/lib/apt/lists/*
+
+WORKDIR /app
+
+COPY requirements.txt .
+
+RUN pip3 install --no-cache-dir -r requirements.txt
+
+COPY . .
+
+EXPOSE 5000
+
+# 启动 whisper_fastapi.py
+ENTRYPOINT ["python3", "whisper_fastapi.py"]

start-docker.sh

@@ -0,0 +1,10 @@
+#!/bin/bash
+
+docker run -d --name whisper-fastapi \
+    --restart unless-stopped \
+    --name whisper-fastapi \
+    -v ~/.cache/huggingface:/root/.cache/huggingface \
+    --gpus all \
+    -p 5000:5000 \
+    docker.io/heimoshuiyu/whisper-fastapi:lastet \
+    --model large-v2

whisper_fastapi.py

@@ -39,7 +39,7 @@ app = FastAPI()
 Instrumentator().instrument(app).expose(app, endpoint="/konele/metrics")
 ccc = opencc.OpenCC("t2s.json")
 
-print("Loading model...")
+print(f"Loading model to device {args.device}...")
 transcriber = Transcribe(
     model_path=args.model,
     device=args.device,
@@ -49,7 +49,7 @@ transcriber = Transcribe(
     cache_directory=args.cache_dir,
     local_files_only=args.local_files_only,
 )
-print("Model loaded!")
+print(f"Model loaded to device {transcriber.model.model.device}")
 
 
 # allow all cors