fix docker write error

DOCUMENTATION.md

@@ -1,12 +1,12 @@
-# Enhanced Multilingual Audio Intelligence System - Technical Documentation
+# Multilingual Audio Intelligence System - Technical Documentation
 
 ## 1. Project Overview
 
-The Enhanced Multilingual Audio Intelligence System is an AI-powered platform that combines speaker diarization, automatic speech recognition, and neural machine translation to deliver comprehensive audio analysis capabilities. This system processes multilingual audio content with support for Indian languages, identifies individual speakers, transcribes speech with high accuracy, and provides translations across 100+ languages through a multi-tier fallback system, transforming raw audio into structured, actionable insights.
+The Multilingual Audio Intelligence System is an AI-powered platform that combines speaker diarization, automatic speech recognition, and neural machine translation to deliver comprehensive audio analysis capabilities. This system processes multilingual audio content with support for Indian languages, identifies individual speakers, transcribes speech with high accuracy, and provides translations across 100+ languages through a multi-tier fallback system, transforming raw audio into structured, actionable insights.
 
 ## 2. Objective
 
-The primary objective of the Enhanced Multilingual Audio Intelligence System is to provide comprehensive audio content analysis capabilities by:
+The primary objective of the Multilingual Audio Intelligence System is to provide comprehensive audio content analysis capabilities by:
 
 - **Language Support**: Support for Tamil, Hindi, Telugu, Gujarati, Kannada, and other regional languages
 - **Multi-Tier Translation**: Fallback system ensuring broad translation coverage across language pairs
@@ -180,7 +180,7 @@ The application includes a demo mode for testing without waiting for full model
 - Available demos:
   - [Yuri_Kizaki.mp3](https://www.mitsue.co.jp/service/audio_and_video/audio_production/media/narrators_sample/yuri_kizaki/03.mp3) — Japanese narration about website communication
   - [Film_Podcast.mp3](https://www.lightbulblanguages.co.uk/resources/audio/film-podcast.mp3) — French podcast discussing films like The Social Network
-  - [Tamil_Wikipedia_Interview.ogg](https://commons.wikimedia.org/wiki/File:Tamil_Wikipedia_Interview.ogg) — Tamil language interview (36+ minutes)
+  - [Tamil_Wikipedia_Interview.ogg](https://commons.wikimedia.org/wiki/File:Parvathisri-Wikipedia-Interview-Vanavil-fm.ogg) — Tamil language interview (36+ minutes)
   - [Car_Trouble.mp3](https://www.tuttlepublishing.com/content/docs/9780804844383/06-18%20Part2%20Car%20Trouble.mp3) — Conversation about waiting for a mechanic and basic assistance (2:45)
 - Static serving: demo audio is exposed at `/demo_audio/<filename>` for local preview.
 - The UI provides enhanced selectable cards under Demo Mode; once selected, the system loads a preview and renders a waveform using HTML5 Canvas (Web Audio API) before processing.

Dockerfile

@@ -24,8 +24,12 @@ RUN pip install --no-cache-dir --upgrade pip && \
 COPY . .
 
 # Create necessary directories with proper permissions
+# Fixed: Use 777 permissions for directories that need write access
 RUN mkdir -p templates static uploads outputs model_cache temp_files demo_results demo_audio \
-    && chmod -R 755 templates static uploads outputs model_cache temp_files demo_results demo_audio
+    /tmp/matplotlib /tmp/fontconfig \
+    && chmod -R 777 templates static \
+    && chmod -R 777 uploads outputs model_cache temp_files demo_results demo_audio \
+    && chmod -R 777 /tmp/matplotlib /tmp/fontconfig
 
 # Set environment variables for Hugging Face Spaces
 ENV PYTHONPATH=/app \
@@ -44,7 +48,8 @@ ENV PYTHONPATH=/app \
     PYANNOTE_CACHE=/app/model_cache \
     MPLCONFIGDIR=/tmp/matplotlib \
     HUGGINGFACE_HUB_CACHE=/app/model_cache \
-    HF_HUB_CACHE=/app/model_cache
+    HF_HUB_CACHE=/app/model_cache \
+    FONTCONFIG_PATH=/tmp/fontconfig
 
 # Expose port for Hugging Face Spaces
 EXPOSE 7860
@@ -54,4 +59,16 @@ HEALTHCHECK --interval=30s --timeout=30s --start-period=60s --retries=3 \
     CMD curl -f http://localhost:7860/api/system-info || exit 1
 
 # Preload models and start the application
-CMD ["python", "-c", "import subprocess; import time; print('🚀 Starting Enhanced Multilingual Audio Intelligence System...'); subprocess.run(['python', 'model_preloader.py']); print('✅ Models loaded successfully'); import uvicorn; uvicorn.run('web_app:app', host='0.0.0.0', port=7860, workers=1, log_level='info')"]
+# Fixed: Ensure directories exist with proper permissions at runtime
+CMD ["python", "-c", "\
+import os; \
+import subprocess; \
+import time; \
+print('🚀 Starting Multilingual Audio Intelligence System...'); \
+for dir in ['uploads', 'outputs', 'model_cache', 'temp_files', 'demo_results', '/tmp/matplotlib', '/tmp/fontconfig']: \
+    os.makedirs(dir, mode=0o777, exist_ok=True); \
+subprocess.run(['python', 'model_preloader.py']); \
+print('✅ Models loaded successfully'); \
+import uvicorn; \
+uvicorn.run('web_app:app', host='0.0.0.0', port=7860, workers=1, log_level='info')\
+"]

README.md

@@ -1,5 +1,5 @@
 ---
-title: Enhanced Multilingual Audio Intelligence System
+title: Multilingual Audio Intelligence System
 emoji: 🎵
 colorFrom: blue
 colorTo: purple
@@ -8,7 +8,7 @@ pinned: false
 short_description: AI for multilingual transcription & Indian language support
 ---
 
-# 🎵 Enhanced Multilingual Audio Intelligence System
+# 🎵 Multilingual Audio Intelligence System
 
 <p align="center">
   <img src="static/imgs/banner.png" alt="Multilingual Audio Intelligence System Banner" style="border: 1px solid black"/>
@@ -48,10 +48,10 @@ This AI-powered platform combines speaker diarization, automatic speech recognit
 
 The system includes sample audio files for testing and demonstration:
 
-- **Japanese Business Audio**: Professional voice message about website communication
-- **French Film Podcast**: Discussion about movies including Social Network and Paranormal Activity
-- **Tamil Wikipedia Interview**: Tamil language interview on collaborative knowledge sharing (36+ minutes)
-- **Hindi Car Trouble**: Hindi conversation about daily life scenarios (2:45)
+- [Japanese Business Audio](https://www.mitsue.co.jp/service/audio_and_video/audio_production/media/narrators_sample/yuri_kizaki/03.mp3): Professional voice message about website communication
+- [French Film Podcast](https://www.lightbulblanguages.co.uk/resources/audio/film-podcast.mp3): Discussion about movies including Social Network and Paranormal Activity
+- [Tamil Wikipedia Interview](https://commons.wikimedia.org/wiki/File:Parvathisri-Wikipedia-Interview-Vanavil-fm.ogg): Tamil language interview on collaborative knowledge sharing (36+ minutes)
+- [Hindi Car Trouble](https://www.tuttlepublishing.com/content/docs/9780804844383/06-18%20Part2%20Car%20Trouble.mp3): Hindi conversation about daily life scenarios (2:45)
 
 ### Demo Features
 - **Pre-processed Results**: Cached processing for quick demonstration
@@ -111,7 +111,7 @@ The system includes sample audio files for testing and demonstration:
 ### **1. Environment Setup**
 ```bash
 # Clone the enhanced repository
-git clone https://github.com/YourUsername/Enhanced-Multilingual-Audio-Intelligence-System.git
+git clone https://github.com/Prathameshv07/Multilingual-Audio-Intelligence-System.git
 cd Enhanced-Multilingual-Audio-Intelligence-System
 
 # Create conda environment (recommended)
@@ -153,34 +153,34 @@ python run_app.py --mode test    # System testing
 
 ```
 Enhanced-Multilingual-Audio-Intelligence-System/
-├── run_app.py                      # 🆕 Single entry point for all modes
-├── web_app.py                      # Enhanced FastAPI application
-├── src/                            # 🆕 Organized source modules
-│   ├── main.py                     # Enhanced pipeline orchestrator
-│   ├── audio_processor.py          # Enhanced with smart file management
-│   ├── speaker_diarizer.py         # pyannote.audio integration
-│   ├── speech_recognizer.py        # faster-whisper integration
-│   ├── translator.py               # 🆕 3-tier hybrid translation system
-│   ├── output_formatter.py         # Multi-format output generation
-│   ├── demo_manager.py             # Enhanced demo file management
-│   ├── ui_components.py            # Interactive UI components
-│   └── utils.py                    # Enhanced utility functions
-├── demo_audio/                     # Enhanced demo files
-│   ├── Yuri_Kizaki.mp3            # Japanese business communication
-│   ├── Film_Podcast.mp3            # French cinema discussion
-│   ├── Tamil_Wikipedia_Interview.ogg  # 🆕 Tamil language interview
-│   └── Car_Trouble.mp3             # 🆕 Hindi daily conversation
+├── run_app.py                         # Single entry point for all modes
+├── web_app.py                         # Enhanced FastAPI application
+├── src/                               # Organized source modules
+│   ├── main.py                        # Enhanced pipeline orchestrator
+│   ├── audio_processor.py             # Enhanced with smart file management
+│   ├── speaker_diarizer.py            # pyannote.audio integration
+│   ├── speech_recognizer.py           # faster-whisper integration
+│   ├── translator.py                  # 3-tier hybrid translation system
+│   ├── output_formatter.py            # Multi-format output generation
+│   ├── demo_manager.py                # Enhanced demo file management
+│   ├── ui_components.py               # Interactive UI components
+│   └── utils.py                       # Enhanced utility functions
+├── demo_audio/                        # Enhanced demo files
+│   ├── Yuri_Kizaki.mp3                # Japanese business communication
+│   ├── Film_Podcast.mp3               # French cinema discussion
+│   ├── Tamil_Wikipedia_Interview.ogg  # Tamil language interview
+│   └── Car_Trouble.mp3                # Hindi daily conversation
 ├── templates/
-│   └── index.html                  # Enhanced UI with Indian language support
+│   └── index.html                     # Enhanced UI with Indian language support
 ├── static/
-│   └── imgs/                       # Enhanced screenshots and assets
-├── model_cache/                    # Intelligent model caching
-├── outputs/                        # Processing results
-├── requirements.txt                # Enhanced dependencies
-├── README.md                       # This enhanced documentation
-├── DOCUMENTATION.md                # 🆕 Comprehensive technical docs
-├── TECHNICAL_UNDERSTANDING.md      # 🆕 System architecture guide
-└── files_which_are_not_needed/     # 🆕 Archived legacy files
+│   └── imgs/                          # Enhanced screenshots and assets
+├── model_cache/                       # Intelligent model caching
+├── outputs/                           # Processing results
+├── requirements.txt                   # Enhanced dependencies
+├── README.md                          # This enhanced documentation
+├── DOCUMENTATION.md                   # Comprehensive technical docs
+├── TECHNICAL_UNDERSTANDING.md         # System architecture guide
+└── files_which_are_not_needed/        # Archived legacy files
 ```
 
 ## 🌟 Enhanced Usage Examples
@@ -246,23 +246,6 @@ MAX_FILE_SIZE_MB=200                       # Smart file size limit
 - **Device Selection**: CPU (recommended), CUDA (if available)
 - **Cache Management**: Automatic model caching and cleanup
 
-## Problem Statement 6 Alignment
-
-This system addresses **PS-6: "Language-Agnostic Speaker Identification/Verification & Diarization; and subsequent Transcription & Translation System"** with the following capabilities:
-
-### **Current Implementation (70% Coverage)**
-- ✅ **Speaker Diarization**: pyannote.audio for "who spoke when" analysis
-- ✅ **Multilingual ASR**: faster-whisper with automatic language detection
-- ✅ **Neural Translation**: Multi-tier system for 100+ languages
-- ✅ **Audio Format Support**: WAV, MP3, OGG, FLAC, M4A
-- ✅ **User Interface**: Transcripts, visualizations, and translations
-
-### **Enhanced Features (95% Complete)**
-- ✅ **Advanced Speaker Verification**: Multi-model speaker identification with SpeechBrain, Wav2Vec2, and enhanced feature extraction
-- ✅ **Advanced Noise Reduction**: ML-based enhancement with Sepformer, Demucs, and advanced signal processing
-- ✅ **Enhanced Code-switching**: Improved support for mixed language audio with context awareness
-- ✅ **Performance Optimization**: Real-time processing with advanced caching and optimization
-
 ## System Advantages
 
 ### **Reliability**
@@ -335,7 +318,7 @@ docker run -p 8000:7860 audio-intelligence
 ### **Hugging Face Spaces**
 ```yaml
 # spaces.yaml
-title: Enhanced Multilingual Audio Intelligence System
+title: Multilingual Audio Intelligence System
 emoji: 🎵
 colorFrom: blue
 colorTo: purple
@@ -368,4 +351,4 @@ This enhanced system is released under MIT License - see the [LICENSE](LICENSE)
 
 ---
 
-**A comprehensive solution for multilingual audio analysis and translation, designed to handle diverse language requirements and processing scenarios.**
+**A comprehensive solution for multilingual audio analysis and translation, designed to handle diverse language requirements and processing scenarios.**

TECHNICAL_UNDERSTANDING.md

@@ -1,8 +1,8 @@
-# Technical Understanding - Enhanced Multilingual Audio Intelligence System
+# Technical Understanding - Multilingual Audio Intelligence System
 
 ## Architecture Overview
 
-This document provides technical insights into the enhanced multilingual audio intelligence system, designed to address comprehensive audio analysis requirements. The system incorporates **Indian language support**, **multi-tier translation**, **waveform visualization**, and **optimized performance** for various deployment scenarios.
+This document provides technical insights into the multilingual audio intelligence system, designed to address comprehensive audio analysis requirements. The system incorporates **Indian language support**, **multi-tier translation**, **waveform visualization**, and **optimized performance** for various deployment scenarios.
 
 ## System Architecture
 

spaces.yaml

@@ -1,4 +1,4 @@
-title: Enhanced Multilingual Audio Intelligence System
+title: Multilingual Audio Intelligence System
 emoji: 🎵
 colorFrom: blue
 colorTo: purple