Refactor Dockerfile for improved dependency management and enhance Streamlit app initialization with better error handling for language identification and accent classification models

Dockerfile

@@ -1,83 +1,47 @@
-FROM python:3.9
+FROM python:3.9-slim
 
-WORKDIR /app
-
-# Set environment variables to avoid permission errors
-ENV HOME=/app \
-    PYTHONUNBUFFERED=1 \
-    PYTHONDONTWRITEBYTECODE=1 \
-    MPLCONFIGDIR=/tmp/matplotlib \
-    TRANSFORMERS_CACHE=/app/.cache/huggingface \
-    XDG_CACHE_HOME=/app/.cache
+# Set environment variables
+ENV PYTHONUNBUFFERED=1 `
+    PYTHONDONTWRITEBYTECODE=1 `
+    MPLCONFIGDIR=/tmp/matplotlib
 
-# Create non-root user for better security
-RUN groupadd -g 1000 appuser && \
-    useradd -u 1000 -g appuser -s /bin/sh -m appuser
+WORKDIR /app
 
-# Install system dependencies including ffmpeg for audio processing
-# Add retry logic and better mirrors for improved network reliability
-RUN apt-get update --allow-releaseinfo-change || (sleep 2 && apt-get update) && \
-    apt-get install -y \
-    build-essential \
-    curl \
-    software-properties-common \
-    git \
-    ffmpeg \
-    libsndfile1 \
-    ca-certificates \
-    && apt-get clean \
+# Install system dependencies
+RUN apt-get update && `
+    apt-get install -y --no-install-recommends `
+    build-essential `
+    curl `
+    git `
+    ffmpeg `
+    libsndfile1 `
+    && apt-get clean `
     && rm -rf /var/lib/apt/lists/*
 
-# Set pip to have more retries and timeout
-ENV PIP_DEFAULT_TIMEOUT=100
-ENV PIP_RETRIES=3
-
-# Copy requirements and install Python dependencies
-COPY requirements.txt ./
+# Create necessary directories
+RUN mkdir -p /app/tmp_model /tmp/matplotlib
 
-# First install torch and torchaudio separately for better compatibility
-RUN pip install --upgrade pip && \
-    pip install torch==2.0.1 torchaudio==2.0.2 --index-url https://download.pytorch.org/whl/cpu
+# Copy requirements first (for better caching)
+COPY requirements.txt .
 
-# Then install the rest of the requirements with retries
-RUN pip install --no-cache-dir -r requirements.txt || \
-    (sleep 2 && pip install --no-cache-dir -r requirements.txt) || \
-    (sleep 5 && pip install --no-cache-dir -r requirements.txt --use-deprecated=legacy-resolver)
-
-# Install SpeechBrain directly using Git for better compatibility
-RUN pip install git+https://github.com/speechbrain/speechbrain@v0.5.14
+# Install Python dependencies with specific order for compatibility
+RUN pip install --no-cache-dir --upgrade pip && `
+    pip install --no-cache-dir torch==2.0.1 torchaudio==2.0.2 && `
+    pip install --no-cache-dir -r requirements.txt && `
+    pip install --no-cache-dir git+https://github.com/speechbrain/speechbrain.git@v0.5.14
 
 # Copy source code
 COPY src/ ./src/
 
-# Create directories with proper permissions
-RUN mkdir -p /app/cookies /app/tmp_model /app/.streamlit /app/.cache && \
-    chown -R appuser:appuser /app
-
-# Switch to non-root user for better security
-USER appuser
-
-# Create a Streamlit configuration file to avoid permission issues
-RUN mkdir -p /app/.streamlit && \
-    echo '\
-[server]\n\
-port = 8501\n\
-address = "0.0.0.0"\n\
-headless = true\n\
-enableCORS = false\n\
-\n\
-[browser]\n\
-gatherUsageStats = false\n\
-\n\
-[runner]\n\
-fastReruns = true\n\
-' > /app/.streamlit/config.toml
+# Set up Streamlit configuration
+RUN mkdir -p .streamlit && `
+    echo "[server]`nport = 8501`naddress = \"0.0.0.0\"`nheadless = true`n`n[browser]`ngatherUsageStats = false`n`n[runner]`nfastReruns = true" > ./.streamlit/config.toml
 
-# Expose the port Streamlit will run on
+# Expose port
 EXPOSE 8501
 
-# Health check to ensure the service is running
+# Health check
 HEALTHCHECK CMD curl --fail http://localhost:8501/_stcore/health
 
-# Run the Streamlit app
-ENTRYPOINT ["streamlit", "run", "src/streamlit_app.py"]
+# Run the app
+ENTRYPOINT ["streamlit", "run", "src/streamlit_app.py", "--server.port=8501", "--server.address=0.0.0.0"]

src/streamlit_app.py

@@ -139,7 +139,8 @@ def extract_audio(video_path="video.mp4", audio_path="audio.wav"):
 
 class AccentDetector:
     def __init__(self):
-        # Initialize the language identification model
+        # Initialize language identification model
+        self.have_lang_id = False
         try:
             if EncoderClassifier is not None:
                 self.lang_id = EncoderClassifier.from_hparams(
@@ -149,12 +150,10 @@ class AccentDetector:
                 self.have_lang_id = True
             else:
                 st.error("SpeechBrain not available. Language identification disabled.")
-                self.have_lang_id = False
         except Exception as e:
             st.error(f"Error loading language ID model: {str(e)}")
-            self.have_lang_id = False
-          # Initialize the English accent classifier - using VoxLingua107 for now
-        # In production, you'd use a more specialized accent model
+          # Initialize the accent classifier
+        self.have_accent_model = False
         try:
             self.model_name = "speechbrain/lang-id-voxlingua107-ecapa"
             
@@ -162,10 +161,11 @@ class AccentDetector:
             if HAS_AUTO_PROCESSOR:
                 self.processor = AutoProcessor.from_pretrained(self.model_name)
             else:
-                # Fall back to using feature_extractor directly if AutoProcessor is not available
+                # Fall back to using feature_extractor
                 from transformers import AutoFeatureExtractor
                 self.processor = AutoFeatureExtractor.from_pretrained(self.model_name)
-                  self.model = AutoModelForAudioClassification.from_pretrained(self.model_name)
+            
+            self.model = AutoModelForAudioClassification.from_pretrained(self.model_name)
             self.have_accent_model = True
         except Exception as e:
             st.warning(f"Could not load accent model: {str(e)}")