Refactor Dockerfile and Streamlit app to improve dependency installation and update accent detection model

Dockerfile

@@ -2,6 +2,7 @@ FROM python:3.9-slim
 
 WORKDIR /app
 
+# Install system dependencies including ffmpeg for audio processing
 RUN apt-get update && apt-get install -y \
     build-essential \
     curl \
@@ -11,13 +12,21 @@ RUN apt-get update && apt-get install -y \
     libsndfile1 \
     && rm -rf /var/lib/apt/lists/*
 
+# Copy requirements and install Python dependencies
 COPY requirements.txt ./
+RUN pip3 install --no-cache-dir -r requirements.txt
+
+# Copy source code
 COPY src/ ./src/
 
-RUN pip3 install --no-cache-dir -r requirements.txt
+# Create directory for temporary model storage
+RUN mkdir -p /app/tmp_model
 
+# Expose the port Streamlit will run on
 EXPOSE 8501
 
+# Health check to ensure the service is running
 HEALTHCHECK CMD curl --fail http://localhost:8501/_stcore/health
 
+# Run the Streamlit app
 ENTRYPOINT ["streamlit", "run", "src/streamlit_app.py", "--server.port=8501", "--server.address=0.0.0.0"]

src/streamlit_app.py

@@ -5,7 +5,7 @@ import subprocess
 import librosa
 import numpy as np
 import torch
-from speechbrain.pretrained import LanguageIdentification
+from speechbrain.inference.classifiers import EncoderClassifier
 from transformers import AutoProcessor, AutoModelForAudioClassification
 from dotenv import load_dotenv
 import matplotlib.pyplot as plt
@@ -76,7 +76,7 @@ def extract_audio(video_path="video.mp4", audio_path="audio.wav"):
 class AccentDetector:
     def __init__(self):
         # Initialize the language identification model
-        self.lang_id = LanguageIdentification.from_hparams(
+        self.lang_id = EncoderClassifier.from_hparams(
             source="speechbrain/lang-id-commonlanguage_ecapa", 
             savedir="tmp_model"
         )
@@ -96,9 +96,8 @@ class AccentDetector:
         """
         Determine if the speech is English and return confidence score
         """
-        prediction = self.lang_id.classify_file(audio_path)
-        lang = prediction[1]
-        score = float(prediction[0][0])
+        out_prob, score, index, lang = self.lang_id.classify_file(audio_path)
+        score = float(score)
         
         # Check if language is English (slightly fuzzy match)
         is_english = "eng" in lang.lower() or "en-" in lang.lower() or lang.lower() == "en"