Add application file

Dockerfile

@@ -1,13 +1,35 @@
-FROM python:3.9
+FROM debian:bookworm-slim AS builder
 
-RUN useradd -m -u 1000 user
-USER user
-ENV PATH="/home/user/.local/bin:$PATH"
+RUN apt-get update && \
+    apt-get install -y curl git python3 python3-pip python3-venv && \
+    rm -rf /var/lib/apt/lists/* /var/cache/apt/archives/* && \
+    mkdir -p /Whisper-WebUI
 
-WORKDIR /app
+WORKDIR /Whisper-WebUI
 
-COPY --chown=user ./requirements.txt requirements.txt
-RUN pip install --no-cache-dir --upgrade -r requirements.txt
+COPY requirements.txt .
+
+RUN python3 -m venv venv && \
+    . venv/bin/activate && \
+    pip install --no-cache-dir -r requirements.txt
+
+
+FROM debian:bookworm-slim AS runtime
+
+RUN apt-get update && \
+    apt-get install -y curl ffmpeg python3 && \
+    rm -rf /var/lib/apt/lists/* /var/cache/apt/archives/*
+
+WORKDIR /Whisper-WebUI
+
+COPY . .
+COPY --from=builder /Whisper-WebUI/venv /Whisper-WebUI/venv
+
+VOLUME [ "/Whisper-WebUI/models" ]
+VOLUME [ "/Whisper-WebUI/outputs" ]
+
+ENV PATH="/Whisper-WebUI/venv/bin:$PATH"
+ENV LD_LIBRARY_PATH=/Whisper-WebUI/venv/lib64/python3.11/site-packages/nvidia/cublas/lib:/Whisper-WebUI/venv/lib64/python3.11/site-packages/nvidia/cudnn/lib
 
 COPY --chown=user . /app
 CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "7860"]

main.py

@@ -1,10 +1,38 @@
-from fastapi import FastAPI
+from fastapi import FastAPI, File, UploadFile
+from fastapi.responses import FileResponse
+import os
+import shutil
+from modules.whisper.whisper_factory import WhisperFactory
 
 app = FastAPI()
 
-@app.get("/hello")
-def hello():
+# Initialize Whisper inference engine
+whisper_inf = WhisperFactory.create_whisper_inference(
+    whisper_type="faster-whisper",
+    whisper_model_dir=os.path.join("models", "Whisper"),
+    faster_whisper_model_dir=os.path.join("models", "Whisper", "faster-whisper"),
+    insanely_fast_whisper_model_dir=os.path.join("models", "Whisper", "insanely-fast-whisper"),
+    output_dir=os.path.join("outputs"),
+)
+
+@app.post("/upload-video/")
+async def upload_video(file: UploadFile = File(...)):
     """
-    this test hello
+    Upload a video file and get the generated SRT file as a response.
     """
-    return {"hello": "you success deploy"}
+    # Save the uploaded video file temporarily
+    input_video_path = os.path.join("temp", file.filename)
+    os.makedirs("temp", exist_ok=True)
+    
+    with open(input_video_path, "wb") as buffer:
+        shutil.copyfileobj(file.file, buffer)
+
+    # Generate the subtitle file
+    output_srt_path = whisper_inf.transcribe_file(
+        input_video_path,
+        file_format="SRT",
+        add_timestamp=True
+    )
+
+    # Return the SRT file as a response
+    return FileResponse(path=output_srt_path, filename=os.path.basename(output_srt_path))

requirements.txt

@@ -1,2 +1,16 @@
+# Remove the --extra-index-url line below if you're not using Nvidia GPU.
+# If you're using it, update url to your CUDA version (CUDA 12.1 is minimum requirement):
+# For CUDA 12.1, use : https://download.pytorch.org/whl/cu121
+# For CUDA 12.4, use : https://download.pytorch.org/whl/cu124
+--extra-index-url https://download.pytorch.org/whl/cu124
+
+torch
+git+https://github.com/jhj0517/jhj0517-whisper.git
+faster-whisper==1.0.3
+transformers==4.42.3
+gradio==4.29.0
+pytubefix
+pyannote.audio==3.3.1
+
 fastapi
 uvicorn