Initial release: DictIA v0.8.14-alpha (fork de Speakr, AGPL-3.0)

config/docker-compose.example.yml

@@ -0,0 +1,65 @@
+services:
+  app:
+    # Use 'lite' tag for a smaller image (~700MB vs ~4.4GB) without PyTorch
+    # Semantic search in Inquire Mode falls back to text search; all other features work normally
+    image: dictia:latest
+    container_name: dictia
+    restart: unless-stopped
+    ports:
+      - "8899:8899"
+
+    # --- Configuration ---
+    # Environment variables are loaded from the .env file.
+    #
+    # To get started:
+    # 1. Copy this file to your project root:
+    #    cp config/docker-compose.example.yml docker-compose.yml
+    #
+    # 2. Copy the unified transcription config (RECOMMENDED):
+    #    cp config/env.transcription.example .env
+    #
+    #    This supports all providers with auto-detection:
+    #    - OpenAI GPT-4o with diarization (set TRANSCRIPTION_MODEL=gpt-4o-transcribe-diarize)
+    #    - Self-hosted ASR/WhisperX (set ASR_BASE_URL=http://your-asr:9000)
+    #    - Legacy Whisper (set TRANSCRIPTION_MODEL=whisper-1)
+    #
+    #    Legacy config files (still supported):
+    #    - config/env.whisper.example    - Standard Whisper API
+    #    - config/env.whisperx.example   - WhisperX with voice profiles
+    #    - config/env.asr.example        - Basic ASR with diarization
+    #
+    # 3. Edit the .env file to add your API keys:
+    #    - TRANSCRIPTION_API_KEY (for OpenAI) or ASR_BASE_URL (for self-hosted)
+    #    - TEXT_MODEL_API_KEY (REQUIRED for summaries, titles, and chat)
+    #
+    # 4. Start DictIA:
+    #    docker compose up -d
+    env_file:
+      - .env
+    
+    environment:
+      # Set log level for troubleshooting
+      # Use ERROR for production (minimal logs)
+      # Use INFO for debugging issues (recommended when troubleshooting)
+      # Use DEBUG for detailed development logging
+      - LOG_LEVEL=ERROR
+
+    # --- Volume Configuration ---
+    # Choose ONE of the following volume configurations.
+    # Option 1 (Recommended): Bind mounts to local folders.
+    volumes:
+      - ./uploads:/data/uploads
+      - ./instance:/data/instance
+      # Optional: Uncomment if using auto-export feature (ENABLE_AUTO_EXPORT=true)
+      # - ./exports:/data/exports
+      # Optional: Uncomment if using auto-processing feature (ENABLE_AUTO_PROCESSING=true)
+      # - ./auto-process:/data/auto-process
+
+    # Option 2: Docker-managed volumes.
+    # volumes:
+    #   - dictia-uploads:/data/uploads
+    #   - dictia-instance:/data/instance
+    #   # Optional: Uncomment if using auto-export feature
+    #   # - dictia-exports:/data/exports
+    #   # Optional: Uncomment if using auto-processing feature
+    #   # - dictia-auto-process:/data/auto-process

config/env.asr.example

@@ -0,0 +1,259 @@
+# -----------------------------------------------------------------------------
+# DictIA Configuration: ASR Endpoint (Legacy)
+#
+# ⚠️  DEPRECATION NOTICE: This configuration style is still supported but
+# we recommend using the new unified configuration in env.transcription.example
+# which supports all transcription providers with auto-detection.
+#
+# Migration: Simply set ASR_BASE_URL and the connector will auto-detect ASR mode.
+# USE_ASR_ENDPOINT=true is no longer required (but still works for backwards compat).
+#
+# Instructions:
+# 1. Copy this file to a new file named .env
+#    cp env.asr.example .env
+# 2. Fill in the required URLs, API keys, and settings below.
+# -----------------------------------------------------------------------------
+
+# --- Text Generation Model (for summaries, titles, etc.) ---
+TEXT_MODEL_BASE_URL=https://openrouter.ai/api/v1
+TEXT_MODEL_API_KEY=your_openrouter_api_key
+TEXT_MODEL_NAME=openai/gpt-4o-mini
+
+# --- GPT-5 Specific Settings (only used with OpenAI API and GPT-5 models) ---
+# If using GPT-5 models (gpt-5, gpt-5-mini, gpt-5-nano, gpt-5-chat-latest) with OpenAI API,
+# these parameters will be used instead of temperature.
+#
+# Example GPT-5 configuration:
+# TEXT_MODEL_BASE_URL=https://api.openai.com/v1
+# TEXT_MODEL_NAME=gpt-5-mini
+#
+# Reasoning effort: minimal, low, medium, high (default: medium)
+# - minimal: Fastest responses, minimal reasoning tokens
+# - low: Fast responses with basic reasoning
+# - medium: Balanced reasoning and speed (recommended)
+# - high: Maximum reasoning for complex tasks
+GPT5_REASONING_EFFORT=medium
+#
+# Verbosity: low, medium, high (default: medium)
+# - low: Concise responses
+# - medium: Balanced detail
+# - high: Detailed explanations
+GPT5_VERBOSITY=medium
+
+# --- Chat Model Configuration (Optional) ---
+# Configure a separate model for real-time chat interactions.
+# If not set, chat will use the TEXT_MODEL_* settings above.
+#
+# Use cases:
+# - Use a faster model for chat while using a more capable model for summarization
+# - Use a cheaper model for interactive chat to reduce costs
+# - Use different service tiers for different operations
+#
+# CHAT_MODEL_API_KEY=your_chat_api_key
+# CHAT_MODEL_BASE_URL=https://openrouter.ai/api/v1
+# CHAT_MODEL_NAME=openai/gpt-4o
+
+# --- Chat GPT-5 Settings (only used with OpenAI API and GPT-5 chat models) ---
+# These settings allow independent control of GPT-5 parameters for chat.
+# If not set, falls back to the main GPT5_* settings above.
+#
+# CHAT_GPT5_REASONING_EFFORT=medium
+# CHAT_GPT5_VERBOSITY=medium
+
+# --- LLM Streaming Compatibility ---
+# Some LLM servers (e.g., certain vLLM configurations) don't support OpenAI's
+# stream_options parameter. If chat streaming hangs or fails, try disabling this.
+# Note: When disabled, token usage tracking for chat will not be available.
+# ENABLE_STREAM_OPTIONS=false
+
+# --- Transcription Service (ASR Endpoint) ---
+# New connector architecture auto-detects ASR mode when ASR_BASE_URL is set.
+# USE_ASR_ENDPOINT=true is deprecated but still works for backwards compatibility.
+#
+# Note: ASR endpoints handle chunking internally - CHUNK_LIMIT settings are ignored.
+
+# ASR Endpoint URL (setting this auto-enables ASR mode)
+# For containers in same docker-compose: Use container name and internal port
+# Example: http://whisper-asr:9000 (NOT the host port 6002 or external IP)
+# For external ASR: Use http://192.168.1.100:9000 or http://asr.example.com:9000
+ASR_BASE_URL=http://whisper-asr:9000
+
+# Deprecated: No longer needed, kept for backwards compatibility
+# USE_ASR_ENDPOINT=true
+
+# Speaker diarization options
+ASR_DIARIZE=true
+# ASR_MIN_SPEAKERS=1         # Hint for minimum speakers
+# ASR_MAX_SPEAKERS=5         # Hint for maximum speakers
+# ASR_RETURN_SPEAKER_EMBEDDINGS=false  # Only enable for WhisperX ASR service
+
+# --- ASR Chunking (for GPUs with limited memory) ---
+# Self-hosted ASR services may crash on long files due to GPU memory exhaustion.
+# Enable app-level chunking to split long files before sending to ASR.
+# Default: false (ASR service handles files internally)
+# ASR_ENABLE_CHUNKING=true
+
+# Maximum audio duration per chunk in seconds (default: 7200 = 2 hours)
+# Lower this value if your GPU runs out of memory on long files.
+# Common values: 600 (10 min), 1200 (20 min), 1800 (30 min), 3600 (1 hour)
+# ASR_MAX_DURATION_SECONDS=7200
+
+# --- Application Settings ---
+# Set to "true" to allow user registration, "false" to disable
+ALLOW_REGISTRATION=false
+# Comma-separated list of allowed email domains for registration.
+# Leave empty to allow all domains. Example: company.com,subsidiary.org
+REGISTRATION_ALLOWED_DOMAINS=
+SUMMARY_MAX_TOKENS=8000
+CHAT_MAX_TOKENS=5000
+
+# Timezone for displaying dates and times in the UI
+# Use a valid TZ database name (e.g., "America/New_York", "Europe/London", "UTC")
+TIMEZONE="UTC"
+
+# Set the logging level for the application.
+# Options: DEBUG, INFO, WARNING, ERROR
+LOG_LEVEL="INFO"
+
+# --- Audio Compression ---
+# Automatically compress lossless uploads (WAV, AIFF) to save storage
+AUDIO_COMPRESS_UPLOADS=true
+
+# Target codec: mp3 (lossy, smallest), flac (lossless), opus (lossy, efficient)
+AUDIO_CODEC=mp3
+
+# Bitrate for lossy codecs (ignored for FLAC)
+AUDIO_BITRATE=128k
+
+# Unsupported codecs - comma-separated list of codecs to exclude from supported list
+# Use this if your transcription service doesn't support certain codecs
+# Supported codecs by default: pcm_s16le, pcm_s24le, pcm_f32le, mp3, flac, opus, vorbis, aac
+# Example: AUDIO_UNSUPPORTED_CODECS=opus,vorbis
+# AUDIO_UNSUPPORTED_CODECS=
+
+# --- Admin User (created on first run) ---
+ADMIN_USERNAME=admin
+ADMIN_EMAIL=admin@example.com
+ADMIN_PASSWORD=changeme
+
+# --- Inquire Mode (AI search across all recordings) ---
+# Set to "true" to enable semantic search and chat across all recordings
+# Requires additional dependencies (already included in Docker image)
+ENABLE_INQUIRE_MODE=false
+
+# --- Automated File Processing (Black Hole Directory) ---
+# Set to "true" to enable automated file processing
+ENABLE_AUTO_PROCESSING=false
+
+# --- Automated Export Settings ---
+# Automatically export transcriptions and summaries to markdown files
+ENABLE_AUTO_EXPORT=false
+
+# Directory where exports will be saved (per-user subdirectories created automatically)
+AUTO_EXPORT_DIR=/data/exports
+
+# What to include in exports
+AUTO_EXPORT_TRANSCRIPTION=true
+AUTO_EXPORT_SUMMARY=true
+
+# Processing mode: admin_only, user_directories, or single_user
+AUTO_PROCESS_MODE=admin_only
+
+# Directory to watch for new audio files
+AUTO_PROCESS_WATCH_DIR=/data/auto-process
+
+# How often to check for new files (seconds)
+AUTO_PROCESS_CHECK_INTERVAL=30
+
+# How long to wait (seconds) to confirm a file has stopped changing before processing.
+# Increase for slow network transfers (NFS, SMB). Default: 5
+# AUTO_PROCESS_STABILITY_TIME=5
+
+# Default username for single_user mode (only used if AUTO_PROCESS_MODE=single_user)
+# AUTO_PROCESS_DEFAULT_USERNAME=admin
+
+# --- Auto-Deletion & Retention Settings ---
+# Enable automated deletion of old recordings
+ENABLE_AUTO_DELETION=false
+
+# Number of days to retain recordings (0 = disabled)
+# Example: 90 means recordings older than 90 days will be processed
+GLOBAL_RETENTION_DAYS=90
+
+# Deletion mode: 'audio_only' keeps transcription, 'full_recording' deletes everything
+# audio_only: Deletes audio file but keeps transcription/summary/notes (recommended)
+# full_recording: Permanently deletes the entire recording from database
+DELETION_MODE=audio_only
+
+# --- Permission-Based Deletion Controls ---
+# Allow all users to delete their recordings, or restrict to admins only
+# true: All users can delete their own recordings (default)
+# false: Only admins can delete recordings
+USERS_CAN_DELETE=true
+
+# Delete speaker profiles when all their recordings are removed.
+# Default: false (speaker profiles and voice embeddings are preserved)
+# Set to true for privacy-sensitive deployments where biometric voice data
+# should not outlive the recordings it was derived from.
+# DELETE_ORPHANED_SPEAKERS=false
+
+# --- Internal Sharing Settings ---
+# Enable user-to-user sharing of recordings (works independently of groups)
+ENABLE_INTERNAL_SHARING=false
+
+# Show usernames in the UI (when sharing/viewing shared recordings)
+# true: Display usernames throughout the interface
+# false: Hide usernames (users must know each other's usernames to share)
+SHOW_USERNAMES_IN_UI=false
+
+# --- Public Sharing Settings ---
+# Enable creation of public share links (anonymous access)
+# true: Users can create public links to share recordings externally (default)
+# false: Public sharing is disabled globally
+ENABLE_PUBLIC_SHARING=true
+
+# Note: Admins can control public sharing permissions per-user in the admin dashboard
+# even when ENABLE_PUBLIC_SHARING is true
+
+# --- Incognito Mode (HIPAA-friendly) ---
+# Enable incognito mode for privacy-sensitive transcriptions
+# When enabled, users can upload recordings that are:
+# - Processed on the server but NOT saved to the database
+# - Stored only in the browser's sessionStorage (lost when tab closes)
+# - Audio files are immediately deleted after processing
+# Useful for HIPAA compliance or sensitive recordings
+# Default: false (feature hidden)
+ENABLE_INCOGNITO_MODE=false
+
+# Make incognito mode the default for in-app recordings (toggle starts ON)
+INCOGNITO_MODE_DEFAULT=false
+
+# --- Video Retention ---
+# When enabled, uploaded video files keep their video stream for in-browser playback
+# The audio is extracted to a temp file for transcription, then cleaned up
+# Default: false (video uploads extract audio only, video stream is discarded)
+VIDEO_RETENTION=false
+
+# --- Concurrent Uploads ---
+# Maximum number of simultaneous file uploads (default: 3)
+MAX_CONCURRENT_UPLOADS=3
+
+# --- Background Processing Queues ---
+# Separate queues for transcription (slow) and summary (fast) jobs
+# This prevents slow ASR jobs from blocking quick summary generation
+
+# Transcription queue workers (for ASR processing, default: 2)
+JOB_QUEUE_WORKERS=2
+
+# Summary queue workers (for LLM summarization, default: 2)
+SUMMARY_QUEUE_WORKERS=2
+
+# Maximum retry attempts for failed jobs (default: 3)
+JOB_MAX_RETRIES=3
+
+# --- Docker Settings (rarely need to be changed) ---
+# Database URI - SQLite (default) or PostgreSQL
+SQLALCHEMY_DATABASE_URI=sqlite:////data/instance/transcriptions.db
+# For PostgreSQL, use: postgresql://username:password@hostname:5432/database_name
+# Example: postgresql://speakr:password@postgres:5432/speakr
+UPLOAD_FOLDER=/data/uploads

config/env.email.example

@@ -0,0 +1,109 @@
+###############################################################################
+# Email Verification & Password Reset Configuration
+###############################################################################
+
+# Enable email verification for new user registrations.
+# When enabled, new users must verify their email before full access.
+# Default: false
+ENABLE_EMAIL_VERIFICATION=false
+
+# Require email verification to log in.
+# Only effective when ENABLE_EMAIL_VERIFICATION=true.
+# When true, users cannot log in until they verify their email.
+# Default: false
+REQUIRE_EMAIL_VERIFICATION=false
+
+###############################################################################
+# SMTP Configuration
+###############################################################################
+
+# SMTP server hostname (required for email functionality)
+# Examples: smtp.gmail.com, smtp.sendgrid.net, smtp.mailgun.org
+SMTP_HOST=smtp.gmail.com
+
+# SMTP server port
+# Common ports: 587 (TLS/STARTTLS), 465 (SSL), 25 (unencrypted)
+# Default: 587
+SMTP_PORT=587
+
+# SMTP authentication username (usually your email address)
+SMTP_USERNAME=your-email@gmail.com
+
+# SMTP authentication password
+# For Gmail: Use an App Password (not your regular password)
+# https://support.google.com/accounts/answer/185833
+SMTP_PASSWORD=your-app-password
+
+# Use TLS/STARTTLS encryption (recommended for port 587)
+# Default: true
+SMTP_USE_TLS=true
+
+# Use SSL encryption (for port 465)
+# Note: Only enable one of SMTP_USE_TLS or SMTP_USE_SSL
+# Default: false
+SMTP_USE_SSL=false
+
+# Email address that appears in the "From" field
+# Should be a valid email address, ideally matching your domain
+SMTP_FROM_ADDRESS=noreply@yourdomain.com
+
+# Display name that appears alongside the from address
+# Default: Speakr
+SMTP_FROM_NAME=Speakr
+
+###############################################################################
+# Provider-Specific Examples
+###############################################################################
+
+# --- Gmail ---
+# SMTP_HOST=smtp.gmail.com
+# SMTP_PORT=587
+# SMTP_USE_TLS=true
+# SMTP_USERNAME=your-email@gmail.com
+# SMTP_PASSWORD=your-app-password  # Generate at https://myaccount.google.com/apppasswords
+
+# --- SendGrid ---
+# SMTP_HOST=smtp.sendgrid.net
+# SMTP_PORT=587
+# SMTP_USE_TLS=true
+# SMTP_USERNAME=apikey
+# SMTP_PASSWORD=your-sendgrid-api-key
+
+# --- Mailgun ---
+# SMTP_HOST=smtp.mailgun.org
+# SMTP_PORT=587
+# SMTP_USE_TLS=true
+# SMTP_USERNAME=postmaster@your-domain.mailgun.org
+# SMTP_PASSWORD=your-mailgun-password
+
+# --- Amazon SES ---
+# SMTP_HOST=email-smtp.us-east-1.amazonaws.com
+# SMTP_PORT=587
+# SMTP_USE_TLS=true
+# SMTP_USERNAME=your-ses-smtp-username
+# SMTP_PASSWORD=your-ses-smtp-password
+
+# --- Microsoft 365 / Outlook ---
+# SMTP_HOST=smtp.office365.com
+# SMTP_PORT=587
+# SMTP_USE_TLS=true
+# SMTP_USERNAME=your-email@yourdomain.com
+# SMTP_PASSWORD=your-password
+
+###############################################################################
+# Notes
+###############################################################################
+
+# Token Expiry Times:
+# - Email verification links expire after 24 hours
+# - Password reset links expire after 1 hour
+
+# Migration Behavior:
+# - Existing users are automatically marked as email_verified=true
+# - New users (when feature is enabled) start as email_verified=false
+
+# Security Recommendations:
+# - Always use TLS or SSL encryption
+# - Use app-specific passwords when available (Gmail, etc.)
+# - Consider using a dedicated email service (SendGrid, Mailgun, SES)
+# - Set a strong SECRET_KEY in your Flask configuration

config/env.sso.example

@@ -0,0 +1,32 @@
+###############################################################################
+# SSO (OIDC) Authentication
+###############################################################################
+
+# Enable SSO (Single Sign-On) authentication. Requires discovery URL and client credentials.
+ENABLE_SSO=false
+
+# Display name for the provider (shown in UI button)
+SSO_PROVIDER_NAME=Keycloak
+
+# OIDC client credentials
+SSO_CLIENT_ID=speakr
+SSO_CLIENT_SECRET=change-me
+
+# OIDC discovery document URL (well-known endpoint)
+SSO_DISCOVERY_URL=https://keycloak.example.com/realms/master/.well-known/openid-configuration
+
+# Public redirect URI exposed by Speakr (must be registered in the IdP)
+SSO_REDIRECT_URI=https://speakr.example.com/auth/sso/callback
+
+# Auto-registration settings
+# Allow automatic account creation for new users signing in via SSO.
+SSO_AUTO_REGISTER=true
+
+# Comma-separated list of allowed email domains for auto-registration.
+# Leave empty to allow all domains (e.g., example.com,company.org).
+SSO_ALLOWED_DOMAINS=
+
+# Claims used to map user profile fields
+SSO_DEFAULT_USERNAME_CLAIM=preferred_username
+SSO_DEFAULT_NAME_CLAIM=name
+

config/env.transcription.example

@@ -0,0 +1,289 @@
+# =============================================================================
+# Transcription Connector Configuration
+# =============================================================================
+#
+# DictIA supports multiple transcription providers through a connector-based
+# architecture. This file documents all available configuration options.
+#
+# Quick Start (Simplified):
+# 1. For OpenAI with diarization: Set TRANSCRIPTION_MODEL=gpt-4o-transcribe-diarize
+# 2. For self-hosted ASR: Set ASR_BASE_URL=http://your-asr:9000
+# 3. For legacy Whisper: Set TRANSCRIPTION_API_KEY and optionally TRANSCRIPTION_MODEL
+#
+# Auto-Detection Priority:
+# 1. TRANSCRIPTION_CONNECTOR - explicit connector name (if you need full control)
+# 2. ASR_BASE_URL - if set, uses ASR endpoint connector
+# 3. TRANSCRIPTION_MODEL contains 'gpt-4o' - uses OpenAI Transcribe connector
+# 4. Default - uses OpenAI Whisper connector with TRANSCRIPTION_MODEL or whisper-1
+
+# =============================================================================
+# TEXT GENERATION MODEL (REQUIRED for summaries, titles, chat)
+# =============================================================================
+# DictIA uses a text/LLM model for generating summaries, titles, and chat.
+# This is separate from the transcription model (STT).
+#
+# You can use OpenRouter (recommended - access to many models) or direct OpenAI API.
+
+# OpenRouter example (recommended - supports many models):
+TEXT_MODEL_BASE_URL=https://openrouter.ai/api/v1
+TEXT_MODEL_API_KEY=your_openrouter_api_key
+TEXT_MODEL_NAME=openai/gpt-4o-mini
+
+# OpenAI direct example:
+# TEXT_MODEL_BASE_URL=https://api.openai.com/v1
+# TEXT_MODEL_API_KEY=sk-your_openai_api_key
+# TEXT_MODEL_NAME=gpt-4o-mini
+
+# --- GPT-5 Specific Settings (only used with OpenAI API and GPT-5 models) ---
+# Reasoning effort: minimal, low, medium, high (default: medium)
+GPT5_REASONING_EFFORT=medium
+# Verbosity: low, medium, high (default: medium)
+GPT5_VERBOSITY=medium
+
+# --- Chat Model Configuration (Optional) ---
+# Configure a separate model for real-time chat interactions.
+# If not set, chat will use the TEXT_MODEL_* settings above.
+# CHAT_MODEL_API_KEY=your_chat_api_key
+# CHAT_MODEL_BASE_URL=https://openrouter.ai/api/v1
+# CHAT_MODEL_NAME=openai/gpt-4o
+
+# =============================================================================
+# CONNECTOR SELECTION (Auto-detected if not set)
+# =============================================================================
+# Options: openai_whisper, openai_transcribe, asr_endpoint
+# Leave empty to auto-detect based on other settings
+# TRANSCRIPTION_CONNECTOR=
+
+# Feature flag to enable/disable new connector architecture (default: true)
+# Set to false to use legacy code path for troubleshooting
+# USE_NEW_TRANSCRIPTION_ARCHITECTURE=true
+
+# =============================================================================
+# OPENAI CONFIGURATION (Required for openai_whisper and openai_transcribe)
+# =============================================================================
+TRANSCRIPTION_API_KEY=your_openai_api_key
+TRANSCRIPTION_BASE_URL=https://api.openai.com/v1
+
+# Model Selection - determines which connector is used:
+#
+# whisper-1              - Legacy Whisper model, no diarization, $0.006/min
+#                          Supports: srt, vtt, json, verbose_json output formats
+#
+# gpt-4o-transcribe      - High quality transcription, no diarization, $0.006/min
+#                          Better accuracy than whisper-1, accepts prompts
+#
+# gpt-4o-mini-transcribe - Cost-effective option, no diarization, $0.003/min
+#                          Good for high-volume, budget-conscious use
+#
+# gpt-4o-transcribe-diarize - Speaker diarization!, $0.006/min
+#                             Identifies speakers as A, B, C, D...
+#                             Requires chunking_strategy for audio >30s
+#
+TRANSCRIPTION_MODEL=gpt-4o-transcribe-diarize
+
+# Legacy Whisper model name (used when TRANSCRIPTION_MODEL is not set)
+# WHISPER_MODEL=whisper-1
+
+# =============================================================================
+# ASR ENDPOINT CONFIGURATION (For self-hosted whisper services)
+# =============================================================================
+# Note: USE_ASR_ENDPOINT is deprecated. Just set ASR_BASE_URL instead.
+# The connector will auto-detect ASR mode when ASR_BASE_URL is set.
+# USE_ASR_ENDPOINT=true  # Deprecated - kept for backwards compatibility
+
+# Base URL of your ASR service (required if USE_ASR_ENDPOINT=true)
+# Supports: whisper-asr-webservice, WhisperX, and compatible services
+# ASR_BASE_URL=http://whisper-asr:9000
+
+# Request timeout in seconds (default: 1800 = 30 minutes)
+# Increase for very long audio files
+# ASR_TIMEOUT=1800
+
+# Enable speaker diarization (default: true)
+# ASR_DIARIZE=true
+
+# Speaker count hints (optional, helps with diarization accuracy)
+# ASR_MIN_SPEAKERS=1
+# ASR_MAX_SPEAKERS=5
+
+# Return speaker embeddings for speaker identification (WhisperX only)
+# Enables automatic speaker matching across recordings
+# ASR_RETURN_SPEAKER_EMBEDDINGS=false
+
+# =============================================================================
+# CHUNKING CONFIGURATION (For large files)
+# =============================================================================
+# Chunking is now connector-aware with this priority:
+# 1. Connector handles internally (openai_transcribe, asr_endpoint) → No app chunking
+# 2. ENABLE_CHUNKING=false → Disable chunking (only affects openai_whisper)
+# 3. CHUNK_LIMIT set → Use your settings
+# 4. Connector defaults → Use connector's recommended limits
+# 5. App default → 20MB size-based
+#
+# For openai_transcribe/asr_endpoint: These settings are IGNORED (connector handles it)
+# For openai_whisper: These settings control chunking behavior
+
+# ENABLE_CHUNKING=false  # Uncomment to disable chunking for openai_whisper
+
+# Chunk limit - supports size (20MB) or duration (600s, 10m)
+CHUNK_LIMIT=20MB
+
+# Overlap between chunks in seconds (helps with transcription accuracy at boundaries)
+CHUNK_OVERLAP_SECONDS=3
+
+# =============================================================================
+# EXAMPLE CONFIGURATIONS (Simplified)
+# =============================================================================
+#
+# --- OpenAI with Speaker Diarization (Recommended) ---
+# Just two environment variables needed:
+# TRANSCRIPTION_API_KEY=sk-xxx
+# TRANSCRIPTION_MODEL=gpt-4o-transcribe-diarize
+#
+# --- Self-hosted WhisperX (Best for privacy) ---
+# Just one environment variable needed (auto-detects ASR mode):
+# ASR_BASE_URL=http://whisper-asr:9000
+# Optional:
+# ASR_DIARIZE=true
+# ASR_RETURN_SPEAKER_EMBEDDINGS=true
+#
+# --- OpenAI Whisper (Legacy, no diarization) ---
+# TRANSCRIPTION_API_KEY=sk-xxx
+# TRANSCRIPTION_MODEL=whisper-1
+#
+# --- Custom Whisper model (local or compatible endpoint) ---
+# TRANSCRIPTION_API_KEY=not-needed
+# TRANSCRIPTION_BASE_URL=http://localhost:8080/v1
+# TRANSCRIPTION_MODEL=Systran/faster-distil-whisper-large-v3
+
+# =============================================================================
+# APPLICATION SETTINGS
+# =============================================================================
+
+# --- Admin User (created on first run) ---
+ADMIN_USERNAME=admin
+ADMIN_EMAIL=admin@example.com
+ADMIN_PASSWORD=changeme
+
+# --- Registration & Access ---
+ALLOW_REGISTRATION=false
+# Comma-separated list of allowed email domains for registration.
+# Leave empty to allow all domains. Example: company.com,subsidiary.org
+REGISTRATION_ALLOWED_DOMAINS=
+
+# --- Token Limits ---
+SUMMARY_MAX_TOKENS=8000
+CHAT_MAX_TOKENS=5000
+
+# --- Timezone ---
+# Use a valid TZ database name (e.g., "America/New_York", "Europe/London", "UTC")
+TIMEZONE="UTC"
+
+# --- Logging ---
+LOG_LEVEL="INFO"
+
+# =============================================================================
+# AUDIO PROCESSING
+# =============================================================================
+
+# --- Audio Compression ---
+# Automatically compress lossless uploads (WAV, AIFF) to save storage
+AUDIO_COMPRESS_UPLOADS=true
+
+# Target codec: mp3 (lossy, smallest), flac (lossless), opus (lossy, efficient)
+AUDIO_CODEC=mp3
+
+# Bitrate for lossy codecs (ignored for FLAC)
+AUDIO_BITRATE=128k
+
+# Unsupported codecs - comma-separated list of codecs to exclude
+# Example: AUDIO_UNSUPPORTED_CODECS=opus,vorbis
+# AUDIO_UNSUPPORTED_CODECS=
+
+# =============================================================================
+# OPTIONAL FEATURES
+# =============================================================================
+
+# --- Inquire Mode (AI search across all recordings) ---
+ENABLE_INQUIRE_MODE=false
+
+# --- Automated File Processing (Black Hole Directory) ---
+ENABLE_AUTO_PROCESSING=false
+# AUTO_PROCESS_MODE=admin_only
+# AUTO_PROCESS_WATCH_DIR=/data/auto-process
+
+# --- Automated Export ---
+ENABLE_AUTO_EXPORT=false
+# AUTO_EXPORT_DIR=/data/exports
+
+# --- Auto-Deletion & Retention ---
+ENABLE_AUTO_DELETION=false
+# GLOBAL_RETENTION_DAYS=90
+# DELETION_MODE=audio_only
+
+# --- Sharing Settings ---
+ENABLE_INTERNAL_SHARING=false
+ENABLE_PUBLIC_SHARING=true
+# SHOW_USERNAMES_IN_UI=false
+
+# --- Permission Controls ---
+USERS_CAN_DELETE=true
+
+# Delete speaker profiles when all their recordings are removed.
+# Default: false (speaker profiles and voice embeddings are preserved)
+# Set to true for privacy-sensitive deployments where biometric voice data
+# should not outlive the recordings it was derived from.
+# DELETE_ORPHANED_SPEAKERS=false
+
+# --- Video Retention ---
+# When enabled, uploaded video files keep their video stream for in-browser playback
+# The audio is extracted to a temp file for transcription, then cleaned up
+# Default: false (video uploads extract audio only, video stream is discarded)
+VIDEO_RETENTION=false
+
+# --- Video Passthrough to ASR ---
+# Send original video files directly to ASR without extracting audio.
+# Useful for custom ASR backends that handle video internally (e.g., multi-track audio extraction).
+# When enabled, video files bypass audio extraction, codec conversion, and chunking.
+# Only affects video files — audio uploads are processed normally.
+# Default: false
+# VIDEO_PASSTHROUGH_ASR=false
+
+# --- Concurrent Uploads ---
+# Maximum number of simultaneous file uploads (default: 3)
+MAX_CONCURRENT_UPLOADS=3
+
+# =============================================================================
+# BACKGROUND PROCESSING
+# =============================================================================
+
+# Transcription queue workers (default: 2)
+JOB_QUEUE_WORKERS=2
+
+# Summary queue workers (default: 2)
+SUMMARY_QUEUE_WORKERS=2
+
+# Maximum retry attempts for failed jobs (default: 3)
+JOB_MAX_RETRIES=3
+
+# =============================================================================
+# DOCKER/DATABASE SETTINGS
+# =============================================================================
+
+# Database URI - SQLite (default) or PostgreSQL
+SQLALCHEMY_DATABASE_URI=sqlite:////data/instance/transcriptions.db
+# For PostgreSQL: postgresql://username:password@hostname:5432/database_name
+
+UPLOAD_FOLDER=/data/uploads
+
+# =============================================================================
+# FUTURE: Additional Provider Notes
+# =============================================================================
+# The connector architecture is designed to support additional providers.
+# Future connectors may include:
+#
+# - Deepgram: Known for excellent diarization and real-time transcription
+# - AssemblyAI: Strong diarization with speaker labels
+# - Google Cloud Speech-to-Text: Enterprise-grade with speaker diarization
+#
+# To request a new connector, please open an issue on GitHub.

config/env.whisper.example

@@ -0,0 +1,256 @@
+# -----------------------------------------------------------------------------
+# DictIA Configuration: Standard Whisper API (Legacy)
+#
+# ⚠️  DEPRECATION NOTICE: This configuration style is still supported but
+# we recommend using the new unified configuration in env.transcription.example
+# which supports all transcription providers with auto-detection.
+#
+# Migration: See TRANSCRIPTION_CONNECTOR documentation in env.transcription.example
+# For OpenAI Whisper, simply set:
+#   TRANSCRIPTION_API_KEY=your_key
+#   TRANSCRIPTION_MODEL=whisper-1  (or gpt-4o-transcribe-diarize for diarization)
+#
+# Instructions:
+# 1. Copy this file to a new file named .env
+#    cp env.whisper.example .env
+# 2. Fill in the required API keys and settings below.
+# -----------------------------------------------------------------------------
+
+# --- Text Generation Model (for summaries, titles, etc.) ---
+TEXT_MODEL_BASE_URL=https://openrouter.ai/api/v1
+TEXT_MODEL_API_KEY=your_openrouter_api_key
+TEXT_MODEL_NAME=openai/gpt-4o-mini
+
+# --- GPT-5 Specific Settings (only used with OpenAI API and GPT-5 models) ---
+# If using GPT-5 models (gpt-5, gpt-5-mini, gpt-5-nano, gpt-5-chat-latest) with OpenAI API,
+# these parameters will be used instead of temperature.
+#
+# Example GPT-5 configuration:
+# TEXT_MODEL_BASE_URL=https://api.openai.com/v1
+# TEXT_MODEL_NAME=gpt-5-mini
+#
+# Reasoning effort: minimal, low, medium, high (default: medium)
+# - minimal: Fastest responses, minimal reasoning tokens
+# - low: Fast responses with basic reasoning
+# - medium: Balanced reasoning and speed (recommended)
+# - high: Maximum reasoning for complex tasks
+GPT5_REASONING_EFFORT=medium
+#
+# Verbosity: low, medium, high (default: medium)
+# - low: Concise responses
+# - medium: Balanced detail
+# - high: Detailed explanations
+GPT5_VERBOSITY=medium
+
+# --- Chat Model Configuration (Optional) ---
+# Configure a separate model for real-time chat interactions.
+# If not set, chat will use the TEXT_MODEL_* settings above.
+#
+# Use cases:
+# - Use a faster model for chat while using a more capable model for summarization
+# - Use a cheaper model for interactive chat to reduce costs
+# - Use different service tiers for different operations
+#
+# CHAT_MODEL_API_KEY=your_chat_api_key
+# CHAT_MODEL_BASE_URL=https://openrouter.ai/api/v1
+# CHAT_MODEL_NAME=openai/gpt-4o
+
+# --- Chat GPT-5 Settings (only used with OpenAI API and GPT-5 chat models) ---
+# These settings allow independent control of GPT-5 parameters for chat.
+# If not set, falls back to the main GPT5_* settings above.
+#
+# CHAT_GPT5_REASONING_EFFORT=medium
+# CHAT_GPT5_VERBOSITY=medium
+
+# --- LLM Streaming Compatibility ---
+# Some LLM servers (e.g., certain vLLM configurations) don't support OpenAI's
+# stream_options parameter. If chat streaming hangs or fails, try disabling this.
+# Note: When disabled, token usage tracking for chat will not be available.
+# ENABLE_STREAM_OPTIONS=false
+
+# --- Transcription Service (OpenAI Whisper API) ---
+# New connector architecture is enabled by default.
+# Available models:
+#   whisper-1              - Legacy, no diarization
+#   gpt-4o-transcribe      - High quality, no diarization
+#   gpt-4o-mini-transcribe - Cost-effective, no diarization
+#   gpt-4o-transcribe-diarize - Speaker diarization! (recommended)
+TRANSCRIPTION_BASE_URL=https://api.openai.com/v1
+TRANSCRIPTION_API_KEY=your_openai_api_key
+TRANSCRIPTION_MODEL=whisper-1
+
+# Legacy model name (deprecated, use TRANSCRIPTION_MODEL instead)
+# WHISPER_MODEL=whisper-1
+
+# --- Application Settings ---
+# Set to "true" to allow user registration, "false" to disable
+ALLOW_REGISTRATION=false
+# Comma-separated list of allowed email domains for registration.
+# Leave empty to allow all domains. Example: company.com,subsidiary.org
+REGISTRATION_ALLOWED_DOMAINS=
+SUMMARY_MAX_TOKENS=8000
+CHAT_MAX_TOKENS=5000
+
+# Timezone for displaying dates and times in the UI
+# Use a valid TZ database name (e.g., "America/New_York", "Europe/London", "UTC")
+TIMEZONE="UTC"
+
+# Set the logging level for the application.
+# Options: DEBUG, INFO, WARNING, ERROR
+LOG_LEVEL="INFO"
+
+# --- Large File Chunking ---
+# Chunking is now connector-aware:
+# - openai_transcribe/asr_endpoint: Handled internally, these settings ignored
+# - openai_whisper: Uses these settings for files >25MB
+#
+# ENABLE_CHUNKING=false  # Uncomment to disable (only for openai_whisper)
+
+# Chunk limit - supports size (20MB) or duration (600s, 10m)
+CHUNK_LIMIT=20MB
+
+# Overlap between chunks (seconds)
+CHUNK_OVERLAP_SECONDS=3
+
+# --- Audio Compression ---
+# Automatically compress lossless uploads (WAV, AIFF) to save storage
+AUDIO_COMPRESS_UPLOADS=true
+
+# Target codec: mp3 (lossy, smallest), flac (lossless), opus (lossy, efficient)
+AUDIO_CODEC=mp3
+
+# Bitrate for lossy codecs (ignored for FLAC)
+AUDIO_BITRATE=128k
+
+# Unsupported codecs - comma-separated list of codecs to exclude from supported list
+# Use this if your transcription service doesn't support certain codecs
+# Supported codecs by default: pcm_s16le, pcm_s24le, pcm_f32le, mp3, flac, opus, vorbis, aac
+# Example: AUDIO_UNSUPPORTED_CODECS=opus,vorbis
+# AUDIO_UNSUPPORTED_CODECS=
+
+# --- Admin User (created on first run) ---
+ADMIN_USERNAME=admin
+ADMIN_EMAIL=admin@example.com
+ADMIN_PASSWORD=changeme
+
+# --- Inquire Mode (AI search across all recordings) ---
+# Set to "true" to enable semantic search and chat across all recordings
+# Requires additional dependencies (already included in Docker image)
+ENABLE_INQUIRE_MODE=false
+
+# --- Automated File Processing (Black Hole Directory) ---
+# Set to "true" to enable automated file processing
+ENABLE_AUTO_PROCESSING=false
+
+# --- Automated Export Settings ---
+# Automatically export transcriptions and summaries to markdown files
+ENABLE_AUTO_EXPORT=false
+
+# Directory where exports will be saved (per-user subdirectories created automatically)
+AUTO_EXPORT_DIR=/data/exports
+
+# What to include in exports
+AUTO_EXPORT_TRANSCRIPTION=true
+AUTO_EXPORT_SUMMARY=true
+
+# Processing mode: admin_only, user_directories, or single_user
+AUTO_PROCESS_MODE=admin_only
+
+# Directory to watch for new audio files
+AUTO_PROCESS_WATCH_DIR=/data/auto-process
+
+# How often to check for new files (seconds)
+AUTO_PROCESS_CHECK_INTERVAL=30
+
+# How long to wait (seconds) to confirm a file has stopped changing before processing.
+# Increase for slow network transfers (NFS, SMB). Default: 5
+# AUTO_PROCESS_STABILITY_TIME=5
+
+# Default username for single_user mode (only used if AUTO_PROCESS_MODE=single_user)
+# AUTO_PROCESS_DEFAULT_USERNAME=admin
+
+# --- Auto-Deletion & Retention Settings ---
+# Enable automated deletion of old recordings
+ENABLE_AUTO_DELETION=false
+
+# Number of days to retain recordings (0 = disabled)
+# Example: 90 means recordings older than 90 days will be processed
+GLOBAL_RETENTION_DAYS=90
+
+# Deletion mode: 'audio_only' keeps transcription, 'full_recording' deletes everything
+# audio_only: Deletes audio file but keeps transcription/summary/notes (recommended)
+# full_recording: Permanently deletes the entire recording from database
+DELETION_MODE=audio_only
+
+# --- Permission-Based Deletion Controls ---
+# Allow all users to delete their recordings, or restrict to admins only
+# true: All users can delete their own recordings (default)
+# false: Only admins can delete recordings
+USERS_CAN_DELETE=true
+
+# Delete speaker profiles when all their recordings are removed.
+# Default: false (speaker profiles and voice embeddings are preserved)
+# Set to true for privacy-sensitive deployments where biometric voice data
+# should not outlive the recordings it was derived from.
+# DELETE_ORPHANED_SPEAKERS=false
+
+# --- Internal Sharing Settings ---
+# Enable user-to-user sharing of recordings (works independently of groups)
+ENABLE_INTERNAL_SHARING=false
+
+# Show usernames in the UI (when sharing/viewing shared recordings)
+# true: Display usernames throughout the interface
+# false: Hide usernames (users must know each other's usernames to share)
+SHOW_USERNAMES_IN_UI=false
+
+# --- Public Sharing Settings ---
+# Enable creation of public share links (anonymous access)
+# true: Users can create public links to share recordings externally (default)
+# false: Public sharing is disabled globally
+ENABLE_PUBLIC_SHARING=true
+
+# Note: Admins can control public sharing permissions per-user in the admin dashboard
+# even when ENABLE_PUBLIC_SHARING is true
+
+# --- Incognito Mode (HIPAA-friendly) ---
+# Enable incognito mode for privacy-sensitive transcriptions
+# When enabled, users can upload recordings that are:
+# - Processed on the server but NOT saved to the database
+# - Stored only in the browser's sessionStorage (lost when tab closes)
+# - Audio files are immediately deleted after processing
+# Useful for HIPAA compliance or sensitive recordings
+# Default: false (feature hidden)
+ENABLE_INCOGNITO_MODE=false
+
+# Make incognito mode the default for in-app recordings (toggle starts ON)
+INCOGNITO_MODE_DEFAULT=false
+
+# --- Video Retention ---
+# When enabled, uploaded video files keep their video stream for in-browser playback
+# The audio is extracted to a temp file for transcription, then cleaned up
+# Default: false (video uploads extract audio only, video stream is discarded)
+VIDEO_RETENTION=false
+
+# --- Concurrent Uploads ---
+# Maximum number of simultaneous file uploads (default: 3)
+MAX_CONCURRENT_UPLOADS=3
+
+# --- Background Processing Queues ---
+# Separate queues for transcription (slow) and summary (fast) jobs
+# This prevents slow ASR jobs from blocking quick summary generation
+
+# Transcription queue workers (for ASR processing, default: 2)
+JOB_QUEUE_WORKERS=2
+
+# Summary queue workers (for LLM summarization, default: 2)
+SUMMARY_QUEUE_WORKERS=2
+
+# Maximum retry attempts for failed jobs (default: 3)
+JOB_MAX_RETRIES=3
+
+# --- Docker Settings (rarely need to be changed) ---
+# Database URI - SQLite (default) or PostgreSQL
+SQLALCHEMY_DATABASE_URI=sqlite:////data/instance/transcriptions.db
+# For PostgreSQL, use: postgresql://username:password@hostname:5432/database_name
+# Example: postgresql://speakr:password@postgres:5432/speakr
+UPLOAD_FOLDER=/data/uploads

config/env.whisperx.example

@@ -0,0 +1,241 @@
+# -----------------------------------------------------------------------------
+# DictIA Configuration: WhisperX ASR Endpoint (with Voice Profiles)
+#
+# ⚠️  DEPRECATION NOTICE: This configuration style is still supported but
+# we recommend using the new unified configuration in env.transcription.example
+# which supports all transcription providers with auto-detection.
+#
+# Migration: Simply set ASR_BASE_URL and the connector will auto-detect ASR mode.
+# USE_ASR_ENDPOINT=true is no longer required (but still works for backwards compat).
+#
+# This configuration is for use with the WhisperX ASR Service:
+# https://github.com/murtaza-nasir/whisperx-asr-service
+#
+# Features supported:
+# - Speaker diarization with pyannote/speaker-diarization-community-1
+# - Voice profile embeddings (256-dimensional) for speaker recognition
+# - Automatic speaker matching across recordings
+# - Better timestamp alignment between speakers and words
+#
+# Instructions:
+# 1. Copy this file to a new file named .env
+#    cp config/env.whisperx.example .env
+# 2. Fill in the required URLs, API keys, and settings below.
+# 3. Set up WhisperX ASR Service (see installation guide)
+# -----------------------------------------------------------------------------
+
+# --- Text Generation Model (for summaries, titles, etc.) ---
+TEXT_MODEL_BASE_URL=https://openrouter.ai/api/v1
+TEXT_MODEL_API_KEY=your_openrouter_api_key
+TEXT_MODEL_NAME=openai/gpt-4o-mini
+
+# --- GPT-5 Specific Settings (only used with OpenAI API and GPT-5 models) ---
+# If using GPT-5 models (gpt-5, gpt-5-mini, gpt-5-nano, gpt-5-chat-latest) with OpenAI API,
+# these parameters will be used instead of temperature.
+#
+# Example GPT-5 configuration:
+# TEXT_MODEL_BASE_URL=https://api.openai.com/v1
+# TEXT_MODEL_NAME=gpt-5-mini
+#
+# Reasoning effort: minimal, low, medium, high (default: medium)
+# - minimal: Fastest responses, minimal reasoning tokens
+# - low: Fast responses with basic reasoning
+# - medium: Balanced reasoning and speed (recommended)
+# - high: Maximum reasoning for complex tasks
+GPT5_REASONING_EFFORT=medium
+#
+# Verbosity: low, medium, high (default: medium)
+# - low: Concise responses
+# - medium: Balanced detail
+# - high: Detailed explanations
+GPT5_VERBOSITY=medium
+
+# --- Auto-Identify Speaker Response Format ---
+# When enabled, auto-identify uses JSON Schema response format (structured outputs)
+# to constrain LLM output to valid SPEAKER_XX keys. Falls back to json_object mode
+# if the model doesn't support it. Leave disabled for widest model compatibility.
+# AUTO_IDENTIFY_RESPONSE_SCHEMA=1
+
+# --- Chat Model Configuration (Optional) ---
+# Configure a separate model for real-time chat interactions.
+# If not set, chat will use the TEXT_MODEL_* settings above.
+#
+# Use cases:
+# - Use a faster model for chat while using a more capable model for summarization
+# - Use a cheaper model for interactive chat to reduce costs
+# - Use different service tiers for different operations
+#
+# CHAT_MODEL_API_KEY=your_chat_api_key
+# CHAT_MODEL_BASE_URL=https://openrouter.ai/api/v1
+# CHAT_MODEL_NAME=openai/gpt-4o
+
+# --- Chat GPT-5 Settings (only used with OpenAI API and GPT-5 chat models) ---
+# These settings allow independent control of GPT-5 parameters for chat.
+# If not set, falls back to the main GPT5_* settings above.
+#
+# CHAT_GPT5_REASONING_EFFORT=medium
+# CHAT_GPT5_VERBOSITY=medium
+
+# --- Transcription Service (WhisperX ASR Endpoint) ---
+# New connector architecture auto-detects ASR mode when ASR_BASE_URL is set.
+# USE_ASR_ENDPOINT=true is deprecated but still works for backwards compatibility.
+#
+# Note: ASR endpoints handle chunking internally - CHUNK_LIMIT settings are ignored.
+
+# WhisperX ASR Endpoint URL (setting this auto-enables ASR mode)
+# For containers in same docker-compose: Use container name and internal port
+# Example: http://whisperx-asr:9000 (NOT the host port or external IP)
+# For external ASR: Use http://192.168.1.100:9000 or http://asr.example.com:9000
+ASR_BASE_URL=http://whisperx-asr:9000
+
+# Deprecated: No longer needed, kept for backwards compatibility
+# USE_ASR_ENDPOINT=true
+
+# Speaker diarization options
+ASR_DIARIZE=true
+# ASR_MIN_SPEAKERS=1         # Hint for minimum speakers
+# ASR_MAX_SPEAKERS=5         # Default maximum speakers
+
+# Enable speaker embeddings for voice profile matching (WhisperX only)
+ASR_RETURN_SPEAKER_EMBEDDINGS=true
+
+# --- Application Settings ---
+# Set to "true" to allow user registration, "false" to disable
+ALLOW_REGISTRATION=false
+# Comma-separated list of allowed email domains for registration.
+# Leave empty to allow all domains. Example: company.com,subsidiary.org
+REGISTRATION_ALLOWED_DOMAINS=
+SUMMARY_MAX_TOKENS=8000
+CHAT_MAX_TOKENS=5000
+
+# Timezone for displaying dates and times in the UI
+# Use a valid TZ database name (e.g., "America/New_York", "Europe/London", "UTC")
+TIMEZONE="UTC"
+
+# Set the logging level for the application.
+# Options: DEBUG, INFO, WARNING, ERROR
+LOG_LEVEL="INFO"
+
+# --- Audio Compression ---
+# Automatically compress lossless uploads (WAV, AIFF) to save storage
+AUDIO_COMPRESS_UPLOADS=true
+
+# Target codec: mp3 (lossy, smallest), flac (lossless), opus (lossy, efficient)
+AUDIO_CODEC=mp3
+
+# Bitrate for lossy codecs (ignored for FLAC)
+AUDIO_BITRATE=128k
+
+# --- Admin User (created on first run) ---
+ADMIN_USERNAME=admin
+ADMIN_EMAIL=admin@example.com
+ADMIN_PASSWORD=changeme
+
+# --- Inquire Mode (AI search across all recordings) ---
+# Set to "true" to enable semantic search and chat across all recordings
+# Requires additional dependencies (already included in Docker image)
+ENABLE_INQUIRE_MODE=false
+
+# --- Automated File Processing (Black Hole Directory) ---
+# Set to "true" to enable automated file processing
+ENABLE_AUTO_PROCESSING=false
+
+# --- Automated Export Settings ---
+# Automatically export transcriptions and summaries to markdown files
+ENABLE_AUTO_EXPORT=false
+
+# Directory where exports will be saved (per-user subdirectories created automatically)
+AUTO_EXPORT_DIR=/data/exports
+
+# What to include in exports
+AUTO_EXPORT_TRANSCRIPTION=true
+AUTO_EXPORT_SUMMARY=true
+
+# Processing mode: admin_only, user_directories, or single_user
+AUTO_PROCESS_MODE=admin_only
+
+# Directory to watch for new audio files
+AUTO_PROCESS_WATCH_DIR=/data/auto-process
+
+# How often to check for new files (seconds)
+AUTO_PROCESS_CHECK_INTERVAL=30
+
+# How long to wait (seconds) to confirm a file has stopped changing before processing.
+# Increase for slow network transfers (NFS, SMB). Default: 5
+# AUTO_PROCESS_STABILITY_TIME=5
+
+# Default username for single_user mode (only used if AUTO_PROCESS_MODE=single_user)
+# AUTO_PROCESS_DEFAULT_USERNAME=admin
+
+# --- Auto-Deletion & Retention Settings ---
+# Enable automated deletion of old recordings
+ENABLE_AUTO_DELETION=false
+
+# Number of days to retain recordings (0 = disabled)
+# Example: 90 means recordings older than 90 days will be processed
+GLOBAL_RETENTION_DAYS=90
+
+# Deletion mode: 'audio_only' keeps transcription, 'full_recording' deletes everything
+# audio_only: Deletes audio file but keeps transcription/summary/notes (recommended)
+# full_recording: Permanently deletes the entire recording from database
+DELETION_MODE=audio_only
+
+# --- Permission-Based Deletion Controls ---
+# Allow all users to delete their recordings, or restrict to admins only
+# true: All users can delete their own recordings (default)
+# false: Only admins can delete recordings
+USERS_CAN_DELETE=true
+
+# Delete speaker profiles when all their recordings are removed.
+# Default: false (speaker profiles and voice embeddings are preserved)
+# Set to true for privacy-sensitive deployments where biometric voice data
+# should not outlive the recordings it was derived from.
+# DELETE_ORPHANED_SPEAKERS=false
+
+# --- Internal Sharing Settings ---
+# Enable user-to-user sharing of recordings (works independently of groups)
+ENABLE_INTERNAL_SHARING=false
+
+# Show usernames in the UI (when sharing/viewing shared recordings)
+# true: Display usernames throughout the interface
+# false: Hide usernames (users must know each other's usernames to share)
+SHOW_USERNAMES_IN_UI=false
+
+# --- Public Sharing Settings ---
+# Enable creation of public share links (anonymous access)
+# true: Users can create public links to share recordings externally (default)
+# false: Public sharing is disabled globally
+ENABLE_PUBLIC_SHARING=true
+
+# Note: Admins can control public sharing permissions per-user in the admin dashboard
+# even when ENABLE_PUBLIC_SHARING is true
+
+# --- Video Retention ---
+# When enabled, uploaded video files keep their video stream for in-browser playback
+# The audio is extracted to a temp file for transcription, then cleaned up
+# Default: false (video uploads extract audio only, video stream is discarded)
+VIDEO_RETENTION=false
+
+# --- Concurrent Uploads ---
+# Maximum number of simultaneous file uploads (default: 3)
+MAX_CONCURRENT_UPLOADS=3
+
+# --- Background Processing Queues ---
+# Separate queues for transcription (slow) and summary (fast) jobs
+# This prevents slow ASR jobs from blocking quick summary generation
+
+# Transcription queue workers (for ASR processing, default: 2)
+JOB_QUEUE_WORKERS=2
+
+# Summary queue workers (for LLM summarization, default: 2)
+SUMMARY_QUEUE_WORKERS=2
+
+# Maximum retry attempts for failed jobs (default: 3)
+JOB_MAX_RETRIES=3
+
+# --- Docker Settings (rarely need to be changed) ---
+# Database URI - SQLite (default) or PostgreSQL
+SQLALCHEMY_DATABASE_URI=sqlite:////data/instance/transcriptions.db
+# For PostgreSQL, use: postgresql://username:password@hostname:5432/database_name
+# Example: postgresql://speakr:password@postgres:5432/speakr
+UPLOAD_FOLDER=/data/uploads