Voice Table Management
Overview
The Voice Table is the central interface for managing all uploaded audio files (conversations), providing comprehensive visibility into file status, processing results, and detailed information access. It displays conversations in a hierarchical folder structure with real-time status updates.
Table Structure
Conversation Information
- Name: Display name/title of the conversation
- Owner: User who uploaded the file (
user_id
) - Timestamp: Upload date and time
- Duration: Audio length in formatted string (HH:MM:SS format)
- Media Key: Storage identifier for the audio file
- Folder: Current folder location (hierarchical organization)
Processing Status Indicators
- Uploading: File transfer and initial processing in progress
- Transcribing: Audio-to-text conversion active
- Running Flows: Automation rules executing agents
- Completed: All processing finished successfully
- Error: Processing failed (with detailed error information)
File Integrity
- File Checksum: SHA-256 hash for file integrity verification
- Blob Source: Storage system identifier (Azure, S3, local)
- Run Results Reference: UUID linking to processing results
Actions and Operations
More Info Button (ℹ️)
Click the information icon to access comprehensive details:
Transcription Results
- Complete Transcript: Full audio-to-text conversion
- Language Detection: Automatically detected language
- Confidence Metrics: Transcription accuracy indicators
- Processing Metadata: Transcription service details
Agent Processing Results
{
"flows": [
{
"flowId": "agent_001",
"text": "Processed output text",
"timestamp": 1234567890,
"error": null
}
],
"failed_flows": [
{
"flowId": "agent_002",
"error": "Processing timeout",
"timestamp": 1234567890
}
],
"finalOutput": "Combined results from all successful agents"
}
System Metadata
- Processing Duration: Total time from upload to completion
- Automation Rule Applied: Which rule triggered the processing
- Agent Execution Details: Individual agent performance metrics
- Error Stack Traces: Detailed error information for debugging
Folder Management
Hierarchical Organization
- Root Level: Conversations without folder assignment (
folder_id: null
) - Nested Folders: Support for multi-level folder structures
- Folder Navigation: Click folders to expand/collapse contents
- Breadcrumb Trail: Clear navigation path display
Folder Operations
- Create Folder: Add new organizational folders
- Rename Folder: Update folder names
- Move Files: Drag and drop or bulk move operations
- Delete Folders: Remove empty folders (with safety checks)
File Operations
Individual File Actions
- Move to Folder: Change file location in folder hierarchy
- Update Metadata: Modify conversation name and custom information
- Reprocess: Re-run automation rules with updated configurations
- Download Original: Access original uploaded audio file
- Delete: Remove conversation and all associated data
Bulk Operations
- Multi-Select: Select multiple conversations for batch operations
- Bulk Move: Move multiple files to target folder
- Bulk Delete: Remove multiple conversations simultaneously
- Batch Export: Generate reports for selected conversations
Filtering and Search
Advanced Filtering Options
- Status Filter: Show only files in specific processing states
- Owner Filter: Display files by specific users
- Date Range: Filter by upload time period
- Duration Range: Filter by audio length
- Folder Filter: Show contents of specific folders only
- Blob Source: Filter by storage system type
Search Functionality
- Filename Search: Find conversations by name
- Transcript Search: Full-text search within transcriptions
- Metadata Search: Search custom information fields
- Checksum Lookup: Find files by exact checksum match
- Media Key Search: Locate files by storage identifier
Search Examples
# Find conversations about meetings
transcript:meeting
# Search for files uploaded last week
date:2024-01-15..2024-01-22
# Find long conversations
duration:>30:00
# Search in specific folder
folder:"Important Recordings"
Audio Streaming and Access
Quality Options
- High Quality: Original audio fidelity (default)
- Medium Quality: Compressed for faster streaming
- Low Quality: Highly compressed for bandwidth-limited environments
Streaming Features
- Range Requests: Efficient partial content loading
- Progressive Download: Start playback before full download
- Seek Support: Jump to specific timestamps
- Adaptive Bitrate: Automatic quality adjustment based on connection
Access Methods
# Stream audio directly
GET /api/conversations/service/run-results/{runresultsreference}/stream?quality=high
# Get temporary download URL
GET /api/conversations/service/run-results/{runresultsreference}/audio-url
# Download file directly
GET /api/conversations/service/run-results/{runresultsreference}/download
Data Management
Storage Information
- Storage Usage: Per-file and total usage statistics
- Retention Policies: Automatic cleanup of old files
- Backup Status: Backup and recovery information
- Cost Tracking: Storage cost attribution per client
Data Lifecycle
- Upload: File stored with initial metadata
- Processing: Transcription and agent execution
- Active Use: Regular access and streaming
- Archive: Long-term storage with reduced access
- Deletion: Complete removal from all systems
Performance Metrics
- Processing Times: Average time per processing stage
- Success Rates: Percentage of successful transcriptions and agent executions
- Error Patterns: Common failure points and resolutions
- Usage Analytics: Access patterns and popular content
Troubleshooting Common Issues
Stuck Processing States
Files Stuck in "Uploading"
Symptoms: Conversation remains in uploading state Solutions:
- Check network connectivity and file size limits
- Verify storage system availability
- Review file format compatibility
- Check for storage quota limitations
Files Stuck in "Transcribing"
Symptoms: Transcription process doesn't complete Solutions:
- Verify audio quality and format
- Check transcription service availability
- Review audio duration limits
- Confirm language support
Files Stuck in "Running Flows"
Symptoms: Agent execution doesn't complete Solutions:
- Check LangFlow service connectivity
- Verify agent availability and permissions
- Review automation rule configuration
- Check for agent timeout settings
Data Integrity Issues
Missing Transcripts
Symptoms: Empty or incomplete transcription text Solutions:
- Verify audio quality and clarity
- Check for supported audio formats
- Review language detection settings
- Consider manual transcription services
Missing Processing Results
Symptoms: No agent outputs or empty results Solutions:
- Verify automation rules are active
- Check agent configuration and permissions
- Review rule matching criteria (type + origin)
- Examine agent execution logs
Performance Issues
Slow Loading
Symptoms: Voice Table takes long time to load Solutions:
- Implement pagination for large datasets
- Use filtering to reduce visible conversations
- Check database performance and indexing
- Consider archiving old conversations
Streaming Problems
Symptoms: Audio playback issues or buffering Solutions:
- Try lower quality settings
- Check network bandwidth
- Verify file integrity via checksum
- Test with different browsers/devices
Best Practices
Organization
- Consistent Naming: Use descriptive, searchable conversation names
- Folder Structure: Create logical folder hierarchies
- Regular Cleanup: Archive or delete outdated conversations
- Metadata Usage: Populate custom_info with relevant details
Performance Optimization
- Bulk Operations: Use batch operations for multiple files
- Strategic Filtering: Apply filters to reduce data load
- Regular Maintenance: Clean up failed or abandoned uploads
- Monitor Usage: Track storage and processing costs
Security Considerations
- Access Control: Verify conversation ownership before operations
- Data Retention: Follow organizational data retention policies
- Audit Trails: Maintain logs of all file operations
- Secure Sharing: Use temporary URLs for external access