Update Dockerfile

Dockerfile

@@ -1,47 +1,36 @@
-FROM ghcr.io/ggml-org/llama.cpp:full
+FROM python:3.11-slim
 
-# 1. Switch to root to install system dependencies
-USER root
+# Force logs to show up instantly in Hugging Face
+ENV PYTHONUNBUFFERED=1
 
-# 2. The 'full' image is Ubuntu-based but lacks pip and ffmpeg. Install them.
+# Install only the system dependency needed for edge-tts (audio)
 RUN apt-get update && apt-get install -y --no-install-recommends \
-    python3 \
-    python3-pip \
     ffmpeg \
     && rm -rf /var/lib/apt/lists/*
 
-# 3. Hugging Face Spaces REQUIRES running as a non-root user (uid 1000)
+# Set up the Hugging Face user
 RUN useradd -m -u 1000 user
 USER user
-
-# 4. Set environment variables
 ENV PATH="/home/user/.local/bin:$PATH"
 ENV HF_HOME="/home/user/.cache/huggingface"
-# CRITICAL: This disables buffering so logs print instantly in the HF Console
-ENV PYTHONUNBUFFERED=1
 
 WORKDIR /home/user/app
 
-# 5. Install Python dependencies safely.
-# Because this is pure Ubuntu, pip will correctly pull the instant glibc CPU wheel.
-RUN python3 -m pip install --no-cache-dir --upgrade pip && \
-    python3 -m pip install --no-cache-dir \
+# Install standard Python libraries (No llama.cpp, no C++ compiling)
+RUN pip install --no-cache-dir --upgrade pip && \
+    pip install --no-cache-dir \
+    torch==2.6.0+cpu --extra-index-url https://download.pytorch.org/whl/cpu \
+    "transformers>=4.49.0" \
+    "accelerate>=1.3.0" \
+    "gguf>=0.10.0" \
     flask==3.1.1 \
     edge-tts \
     num2words==0.5.14 \
     "huggingface_hub>=0.27.0" \
-    "numpy<2.0" \
-    llama-cpp-python \
-    --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/cpu
+    "numpy<2.0"
 
-# 6. Copy your Python app into the container
 COPY --chown=user:user . .
 
 EXPOSE 7860
 
-# 7. CRITICAL: We MUST clear the base image's entrypoint. 
-# Otherwise, it attempts to run `/app/llama-cli python3 app.py` and crashes.
-ENTRYPOINT[]
-
-# 8. Start the app
-CMD ["python3", "app.py"]
+CMD["python", "app.py"]